IPG Photonics – growth, cash, markets and technology

Company: IPG Photonics Corporation (IPGP)

Disclosure: I’m long IPGP.

Company website: http://www.ipgphotonics.com/

Morningstar quote / Yahoo quote / Seeking Alpha IPGP page

Valuation September 27 2013 – Price $57.33 / Market cap $3 billion / Forward P/E 15.3 to 15.9 (sources vary) / no regular dividend

Trailing twelve month ratios: P/E 19.90, Price/Sales 4.91, Price/Cash Flow 16.40
Price/Book (Most recent quarter) 3.68

Projected EPS Q4 2012 to Q3 2013 = $2.903 to $3.003
Projected P/E Q4 2012 to Q3 2013 = 19.09 to 19.75
(See ‘Valuation’ at end.)



I’ll be quoting from:

IPG Photonics’ CEO Discusses Q2 2013 Results – Earnings Call Transcript“, Jul 30 2013, seekingalpha.com

which I’ll refer to as “the transcript”.

IPG’s SEC filings are on this page on their website, with links for other reports on the left. The presentation pdf “Needham 2nd Annual Advanced Industrial Technologies Conference” and the Prepared Remarks for the Conference Call can be downloaded from the company’s Events & Presentations page.


IPG’s main business is making fiber lasers for industrial applications. The ‘materials processing’ segment includes marking and engraving, cutting and welding, and accounts for 88% of sales, leaving Telecom, Advanced and Medical which make up the other 12%. The company is pushing its laser technology into high power, fine processing, micromachining, semiconductor, low cost replacements for cheap flash lamp pumped lasers, and advanced laser systems. For many years, that technology was fiber laser technology, but IPG have more recently acquired excimer laser and diode pumped solid state laser technology. The company now produces direct diode lasers, where the beam is from the laser diodes instead of using them to optically pump a gain medium such as active fiber. IPG have also acquired Photonic Crystal Fiber technology, described later.

There’s some history and general background including IPG’s vertical integration, in a previous post: IPG Photonics – growth, a miss, opportunity and threat March 10, 2013, mostly near the beginning.


Results were reported on July 30, 2013, and they were generally good:

Revenue Q2 2013 $168.2 millionup 22% from Q2 2012
Revenue Q2 2012 $137.9 million

Gross margin Q2 2013 53.5%down 0.8% from Q2 2012
Gross margin Q2 2012 54.3%

Operating income Q2 2013 $59.9 millionup 11% from Q2 2012
Operating income Q2 2012 $56.4 million

Operating margin Q2 2013 35.6%down 5.3% from Q2 2012
Operating margin Q2 2012 40.9%

Net income Q2 2013 $41.7 millionup 11% from Q2 2012
Net income Q2 2012 $37.7 million (after subtracting $2.1 million attributable to non-controlling interests)

Earnings per share (diluted) Q2 2013 $0.80up 11% from Q2 2012
Earnings per share (diluted) Q2 2012 $0.72

R&D Q2 2013 $10.5 millionup 45% from Q2 2012
R&D Q2 2012 $7.2 million

There’s more detail in this table, including how some costs have risen as a percentage of revenue:

IPG Photonics Q2 2013 Income

Guidance for Q3 2013

Guidance for Q3 in the Q2 2013 earnings call transcript is for EPS of $0.77 to $0.87 per diluted share (or $0.82 plus or minus $0.05). That compares to $0.81 in Q3 2012. The mid point is only slightly above a 1% increase, which is only a fifth of the implied variation of $0.05, and a y.o.y. fall from $0.81 to $0.77 is within guidance.

Expected Q3 revenue is $165 to $175 million, with the $170 million midpoint 9% above Q3 2012.

Outlook for Q3 2013

Solid demand, a small backlog (book-to-bill was over 1), not much change in OpEx, gross margin 53 to 54% (from the Prepared Remarks and the transcript).

The cost of expansion

The 5.3% drop in operating margin is mostly explained by the rise in Research and development costs and General and administrative costs, as a percentage of revenue. The rises faster than revenue are probably a kind of investment. That’s obvious in the case of R&D (see UV lasers below), and from the Management Comments in the quarterly results:

    “Excluding foreign exchange rate gains, operating income grew by 12.6%. While this was lower than the growth in revenue, it reflects our investment in operating expenses to support IPG’s future growth.”

That might not seem consistent with the 10% rise y.o.y of inventories, compared to the 22% rise y.o.y in revenue. In thousands:

Inventories Q2 2013 $154,093 – up 8% from Q1 2013 – up 10% from Q2 2012
Inventories Q1 2013 $142,096
Inventories Q2 2012 $139,618

unless high inventories a year ago were in preparation for the revenue growth. Year-on-year, the ratio inventory / revenue has fallen from 1.01 to 0.92, implying that if revenue over the next year is correlated with inventories, revenue growth will be less than 22%. I don’t expect a tight correlation, and the R&D expense increased through the Mobius acquisition should add to revenue over the next year and beyond.

IPG has relatively long lead times, low turnover and high inventories, which would generally imply some expense ahead of expansion. From the 2012 10-K:

    “Given our vertical integration, rigorous and time-consuming testing procedures for both internally manufactured and externally purchased components and the lead time required to manufacture components used in our finished products, the rate at which we turn inventory historically has been low when compared to our cost of sales. Also, our historical growth rates require investment in inventories to support future sales and enable us to quote short delivery times to our customers, which we believe provides us with a competitive advantage. Furthermore, if there is a disruption to the manufacturing capacity of any of our key technologies, our inventories of components should enable us to continue to build finished products for a reasonable period of time. We believe that we will continue to maintain a relatively high level of inventory compared to our cost of sales.”

Relatively long lead times, low turnover, and high inventories, sound like what they teach you not to do in business school, but as the 10-K explains, there are advantages for IPG’s customers, as products are thoroughly tested and are more likely to be in-stock. The benefits to the customer have a cost to IPG. Continuing from the 10-K:

    “As a result, we expect to have a significant amount of working capital invested in inventory. A reduction in our level of net sales or the rate of growth of our net sales from their current levels would mean that the rate at which we are able to convert our inventory into cash would decrease.”

There’s also likely to be some extra hiring ahead of expansion, with a lag between recruitment and payback. (Danaher Corporation’s ChemTreat subsidiary has an estimated three-year payback period for new sales employees, see “Danaher’s Management Presents at Morgan Stanley Industrials and Autos Conference (Transcript)“, Sep 16 2013, seekingalpha.com.)

The effects of a downturn would be amplified by the high inventory. If it happened after preparation for expansion, inventory would be higher and recruitment would add to costs. The charts further down show the dip in 2007 and the sharp recovery.

Capital expenditure was $16.5 million in the quarter, in line with the $60 to $70 million expected for 2013 (in the 10-K for 2012). It’s a volatile number annually, and for 2010 to 2012 was $32,559, $79,099, $55,257 (in thousands).

From the transcript:

    “…we expect next quarterly year for revenue a year with a very big growth…”

I think that means big growth over the next four quarters. The quote is relatively straightforward, but I don’t understand a lot of what the CEO is transcribed as saying. You might be better than I am at following a webcast.

This article on Seeking Alpha makes some points about growth:

IPG Photonics Looking To Innovation And Integration” by Stephen Simpson, Aug 27 2013. seekingalpha.com

Gross margin and Cost of sales

The 0.8% drop in gross margin is explained by acquisition costs, product mix, and the mix of customers and regions sold to. In my opinion there must also be some impact from lower pricing to encourage sales of some new models and high power lasers. The rise in Cost of sales by 24.17% exceeds the rise in Revenue of 21.93% by 2.24%. That looks worse than the 0.8% gross margin drop. The numbers here are related by using the same variables, and the 2.24% is a different way of measuring the effect of the same factors that affect the gross margin. I didn’t notice any analysts asking about the faster rise in Cost of sales, possibly because analysts are used to looking at margins.

Q3 2013

The EPS guidance, large operating margin fall and maybe the small gross margin fall or the faster rise in Cost of sales, could explain some of the share-price under-performance.

The unexciting guidance might be explained by a continuation of “investment in operating expenses”. ‘OpEx’ is expected to hold steady at the Q2 2013 level, which gives a 16% y.o.y rise:

Total operating expenses (in thousands)
Q3 2013 – $30,047 – up by 16%
Q3 2012 – $25,952
increase $4,095

Opinion about investment and margins

While some excellent companies (the kind Warren Buffet likes) have slow and steady growth with little investment, there is also a case for companies that can absorb high investment and produce a good return on it, which IPG seems capable of. ‘Invest when the company invests’ might be worth considering when the investment seems to be under-appreciated by the market, which may be more likely when much of the investment looks like rising costs rather than capex. I’ve only just invented the phrase and haven’t had time to test the principle, and there are exceptions. In support, Danaher claim their ChemTreat subsidiary’s expansion through sales recruitment is hard for competitors to copy, because the three-year payback would cause temporary margin dilution and too much pressure from investors. This has let ChemTreat grow about 10% pa (mostly organic) in a market growing about 2% to 3% pa. How far the case generalizes towards investors over-weighting margin without considering the context, is a matter of opinion or experience.

In this and the previous Earnings Call transcript (and maybe more), I have the impression that when analysts ask about margin, they give high margin a high priority. IPG’s officers reply by explaining that high revenue is also good. The principle can be explained with simple math, although there are complications in practice. If you half the margin and double the number of items sold, profit stays the same. If you half the margin and the number of items sold triples, profit increases by 50%. Pricing depends on economies of scale and the sensitivity of demand to price. Some good businesses have high margins (sometimes relatively) and inelastic demand, while at the other end, retailers in price-sensitive markets have successfully used the low margin, high turnover and low cost model. IPG are somewhere in the middle but in a good position, with high margins that are also higher than competitors’ margins, combined with price elasticity. This time the CEO told analysts that a margin of 60% to 65% could be achieved for a while, but at the cost of sales growth.

High and stable margins might be seen as evidence of pricing power, but from what I’ve read about pricing power in relation to value-investing, it’s the ability to raise prices with little effect on demand. When IPG price at a lower margin to encourage demand, by implication they don’t have pricing power in that area, but pricing power is not necessary for a good business. The ability to add a big margin on to costs is an excellent substitute. With high gross and operating margins, and growth, a lower margin on some products to encourage take-up is not a negative. It would be different if margins were relentlessly driven lower by competition.

Some markets such as medical diagnostic equipment have had high margins for a long time, but if lasers are different and margins fall through competition, the best defense is to expand into new applications now, even if the expansion is sometimes helped by lower margins.

If you believe that margins are falling due to competitors catching up in fiber, or competing more effectively by whatever method, it’s your money and I can’t actually prove you wrong.

A long look back

I like charts showing the cash that’s gone into investment and the cash produced by operations. These cash flows are balanced by the cash provided by financing activities, and the net effect on the cash balance.

IPG cash from ops and into investment - spread

IPG cash from ops and into investment

These ‘walk’ charts show the same information as the time-series chart above. The format is unconventional but I find it easier to spot patterns in them.

IPG walks for cash from ops and into investment

My interpretation is that over time, IPG have reinvested most of their cash from operations. The fact that over time, more cash has been produced by operations than has been put into investment, plus the generally bigger steps, indicates that the investment has produced good results. Substantially higher investment in 2007 was followed by two years of substantially higher cash from operations, and substantially higher investment in 2011 was followed by a year of substantially higher cash from operations, although the chart doesn’t actually prove any cause-and-effect. To check that the benefit has not been reduced too much by stock compensation, I repeat the charts on a ‘per share’ basis.

IPG walks per share for cash from ops and into investment

This table allows a conservative assessment of the long term growth in cash from operations per share. While the growth rate from 2004 to 2012 is 39% CAGR, 2004 was a low year and 2012 saw 90% growth over 2011. Looking at the columns for 6, 7 and 8 years ahead, the lowest figure is 25.9%, which is the CAGR growth rate from 2005 to 2011. So, even when choosing the worst of the six longest periods in the range 2004 to 2012, the CAGR growth rate in cash from operations per share was 25.9%.

IPG cash from ops growth - spread

Growth often tails off as a company gets bigger. The healthy figures in the bottom three rows, for 2009, 2010 and 2011, show that hasn’t happened to IPG. The numbers here are more variable, as they only span one, two or three years.

The three negative numbers are in a diagonal, and are the result of the downturn in 2007, which was followed by a sharp recovery.

Balance sheet and commitments

All dollar amounts here are in thousands.

As at March 31, 2013:

Cash $355,715
Total current assets $649,371
Property, plant and equipment, net $218,995
(I haven’t bothered with goodwill, intangibles etc.)

Total liabilities $119,033

In the 10-Q, under ’12. COMMITMENTS AND CONTINGENCIES’, only disputes and legal proceedings are mentioned, with an assurance of no material effect.

Taking the figures from ’10. COMMITMENTS AND CONTINGENCIES’ in the 10-K for 2012 (so, as at December 31 2012):

Commitments for Operating Leases $11,774
Contractual Obligations (construction) $8,921
Legal proceedings (no material effect) $0
Total $20,695

The terms of “…employment agreements with certain members of senior management.” are not disclosed, but include ‘defined severance’.

Adding the 2012 commitments ($20,695) to Q2 2013 Total liabilities ($119,033) gives $139,728.
That equals only 39.28% of the $355,715 cash.

Defined benefit pensions used to be a headache for employers in the UK, but I didn’t see evidence of anything like that in the 10-K.

The balance sheet is strong, and not undermined by anything under ‘COMMITMENTS AND CONTINGENCIES’.

Financing activities

All dollar amounts here are in thousands except for per share amounts.

Net cash provided by financing activities, for 2012, 2011, 2010:
$82,087, $31,885, $37,764
Total: $151,736

The 2012 figure includes +$167,928 raised by a public offering, and -$33,353 from a special dividend of $0.65 per share in December 2012.

Given the $355,715 cash after Q2 2013, the $384,053 at the end of 2012, and the healthy cash flow, management seem to be even more paranoid than I am about IPG needing cash if there’s a downturn, although they might have wanted cash for acquisition opportunities.

The borrowing has not eroded the net-cash position, and the public offering has not stopped the ‘per share’ performance from looking good, in the ‘per share’ walk chart or the growth table shown earlier.

It’s not surprising that IPG have cash, given the charts above and the financing.

Customer concentration

From the 10-K for 2012.

    “Net sales derived from the Company’s five largest customers as a percentage of its annual net sales were 16%, 17% and 19% in 2012.”

    “Sales to the Company’s largest customer accounted for 7%, 8% and 7% of its net sales in 2012, 2011 and 2010.”

Relative size

From the 10-K for 2012, under ‘Competition’:

    “Many of our competitors are larger than we are and have substantially greater financial, managerial and technical resources, more extensive distribution and service networks, greater sales and marketing capacity, and larger installed customer bases than we do.”

Competitors such as Mitsubishi Cable Industries and the Scientific-Atlanta division of Cisco Systems, are parts of much bigger companies. However I’ve seen Trumpf and IPG described as the giants of industrial lasers, or something like that. When I looked into Trumpf in February, they were a private company and didn’t break out revenue for their laser division. I haven’t checked since then. There’s a different view of market share in:

Rofin-Sinar Looking To A Cyclical Recovery And Fiber Share Gains” by Stephen Simpson CFA, Sep 19 2013. seekingalpha.com

where Rofin-Sinar Technologies are top of the pecking order, with IPG well down the list.

Guessing the future

A story title claims that by 2018 the market for laser-cutting will be worth $3.77 billion. You can confirm that here, but you aren’t likely to learn much more without buying an expensive report. For some context, IPG’s 2012 Net sales were $562,528,000 in 2012.


One concern I have is the Chinese market, which suffers from chronic oversupply supported by loans to further politicians’ careers. Companies affected can go bankrupt when the politicians can’t supply enough finance, or lose interest in the company. This shouldn’t directly affect non-Chinese companies that produce or sub-contract in China, and no-one will force IPG to take a loan and fill a warehouse, but some of their customers could be affected. IPG have been successful in China, with sales up 54% year-on-year to $58.5 million, and a “strong backlog”. They claim to be diverse there, so a single bankruptcy shouldn’t have too much effect. IPG realize that credit can tighten fast, and keep an eye on the situation. ‘prepayments’ are mentioned in the transcript, and I expect they have a fairly good idea of who will be able to pay and who won’t. I couldn’t say IPG are overdependent on China, as much of the world’s production takes place there.

There’s more about China and U.S. automotive under the ‘Stories’ tab. The first link, (“Reality dawns on the China-growth myth” By: Merryn Somerset Webb, 27/08/2013), reports that the official rise in sales of passenger vehicles in the first half of the year, does not tally with the experience of dealers.

North America

The automotive industry is picking up in the U.S.A., and from the transcript:

    “The materials processing business in North America for automotive and other applications is continuing to perform well.”


Russia was weak due to low demand from telecoms, offset slightly by an improvement in laser-systems. IPG don’t trust anyone else’s distribution network, due to corruption, so they’re building their own, which takes time and investment. Being savvy about that kind of thing can avoid a lot of trouble, but they ought to know about Russia because that’s where IPG started.


There’s a vague but optimistic phrase mentioning Korea and automotive. IPG employed a new General Manager in Korea in Q4 2012, after under-performance. Korea seems receptive to new industrial technology, including power generation from fuel cells, zinc recovery from furnaces, and superconducting fault-current limiters. It will take some time but I won’t be surprised if big orders come from Korean automotive, possibly for seam-steppers.


Europe was described as flat, at $47 million, but from the Management Comments in the results:

    “Weak automotive sales in Germany slightly offset our otherwise solid growth in Europe.”

European auto companies have production in Asia generating orders for IPG’s lasers. There were more orders from some OEMs, and two orders for laser systems with high value-added (meaning a lot of kit apart from the laser). In case anyone missed it: “IPG lands record automotive deal” 03 May 2013, optics.org. The order was for 100 four kilowatt units to a customer in Germany, with delivery expected over a year.

A remark that European companies appreciate the benefits of IPG’s lasers but are reluctant to invest, underlines the probably obvious point that IPG would suffer if it’s main markets were in recession.


Laser cutting sales to OEMs in Turkey have increased.

Geographic breakdown of revenue

From the Needham presentation:

“North & South America 19%, Europe 36%, Asia & Australia 45%” (% of sales by geography)

The percentages seem old as they agree with the 10-K for 2012. My rough estimates for Q2 2013 simply apply half the growth rates below, which is dubious but at least takes some account of the divergent growth.

North & South America 20%, Europe 33%, Asia & Australia 47%

Revenue growth year-on-year:

North America up 34%
Europe – flat
Asia & Australia up 34%

Turkey is presumably in IPG’s Asia & Australia, as it’s in ‘Western Asia’, an area I haven’t seen figures for. IPG’s short Q4 2012 fact sheet has a slice of pie-chart labeled ‘Europe, Russia & CIS’ (with 36%). CIS is the Commonwealth of Independent States (Wikipedia), with ‘Stans’ (Kazakhstan etc.) and other ex-Soviet Union countries.

The 10-K for 2012 gives revenue for ‘Rest of World’ of about $1.7 million or less than a third of one per cent of revenue. That must include South America, as the American category is ‘United States and other North America’.

Breaking in to new markets

    “We haven’t really heard anything I mean I think there is a rumor out there that IPG was going to suddenly drop the price of pulsed lasers by 25% I think. I don’t know where that came from. We clearly are always introducing new product that has lower cost related to it so try and penetrate new applications. We’re looking at completely new designed pulsed lasers to try and get greater market share from very low cost YAG lasers but some of the rumors out there in the market I don’t think are with foundation.” – from the transcript on seekingalpha.com linked to at the top

When IPG design new lasers with a low build-cost to sell into new applications, my guess is that the usual margin will be reduced where necessary to ensure sales, especially when the margin can be increased over time through cost reductions. Although I wrote ‘low build-cost’, there’s a complication because a higher build-cost with the same margin can be a rational purchase if the total cost of ownership isn’t higher, through efficiency, reliability and other factors. If the ‘very low cost YAG lasers’ are flash lamp pumped, they only have 2% or less wall plug efficiency, and the lamps have a relatively short life.

Generally, rumors of cheap new lasers from IPG are likely to cause some competitors’ customers to delay orders, with a chance of buying from IPG when the lasers are available. Rumors of a big price cut on a wide range of IPG’s existing lasers will probably harm IPG more than competitors, through delayed sales. In my opinion, the second kind of rumor has little credibility and therefore won’t have much impact. Outside of extreme circumstances, IPG are unlikely to make deep and wide cuts on existing lasers, as customers would be wary of buying ahead of price cuts, for a long time after. If the rumors had made a material impact on sales, I think it would have been reported. The rumors could be a result of competitive pricing on some new products, and maybe designing for low cost has been very successful.

At the high-power end, IPG are aware that affordability increases adoption and application in new areas. In the auto industry, prices might be shaved in order to sell in volume, partly because big auto-makers are used to squeezing suppliers, although that’s more a psychological argument than an economic one.

Lasers are not always sold into new areas through lower margins or designing for lower cost. Fifty QCW (Quasi-continuous wave) lasers were sold for glass-cutting, described as “a new application developed by one of our customers” and probably in consumer electronics. There’s the potential for laser glass-cutting to expand, but with experimentation rather than an immediate surge. The CEO seems optimistic about applications beyond consumer electronics, but I wish I knew what he meant by ‘contraction’, if that’s what he said:

    “…new from market windows and our employing to contraction of (inaudible) our lasers very suitable…”

If he meant ‘construction’ as in buildings, that widens the area of application. I can guess that cutting window glass to size on a construction site might be easier with lasers, but you’re likely to get a better opinion about the room for improvement from a construction worker.

UV lasers

The Mobius Photonics acquisition of Q1 2013 was mostly engaged in R&D for UV fiber lasers, so the immediate impact is in the 46% increase in R&D. According to the CEO two months ago, UV lasers used for semiconductors were expensive, and he expects good demand next year for IPG’s new generation UV lasers. However, that might depend on the competition not changing the situation.

The acquisition also gets important IP, including a UV-related patent where photonic crystal fiber is used to counter scattering. UV lasers allow more delicate machining, an area that fiber lasers have not been very successful in so far. See “IPG speeds UV laser development with Mobius acquisition” 14 Mar 2013, optics.org

There’s more about photonic crystal fibers under the ‘Supercontinuum’ sub-heading.

The optics.org link above also tells how Coherent, Newport and Mobius between them launched lasers for UV and micromachining at the same show in February.

Mobius was not the first acquisition with UV expertise. From the latest 10-K:

“In 2012, we acquired a business that develops and produces industrial grade UV excimer, solid state and pico-second laser micromachining systems.”

that must be JP Sercel Associates. Also from the 10-K:

    “Through our IPG Microsystems business, we offer industrial grade UV excimer, diode pumped solid state and pico-second laser micromachining systems and materials processing services. Key applications for these systems include advanced laser scribing and lift-off (“LLO”) of light-emitting diodes (“LEDs”), thin film solar scribing, semiconductor, micro-electro-mechanical systems (“MEMS”), research, biomedical and industrial micromachining. IPG Microsystems’ laser systems operate at wavelengths from 157nm to 1,064nm, and are essential to a growing set of today’s industrial micromachining applications.”

UV excimer and diode pumped solid state are two kinds of laser that aren’t fiber. An excimer laser is a kind of gas laser that avoids having atoms or molecules drop to a low energy state, where they can absorb photons. An excimer molecule has enough energy to emit a photon, but it disintegrates when the photon goes, leaving bits that won’t absorb in the frequency the excimer emits. (Raman lasers and free electron lasers, or FELs, also avoid the resonant low energy state. See under ‘Raman lasers’ and ‘War’.)

Specification diversity

IPG are the leaders in fiber lasers, but little Mobius was ahead of them, launching a fiber UV laser in February. I believe that reflects intense research by many players over many fronts. A laser’s specification includes wavelength, power, efficiency, beam quality, reliability, footprint, the shape of the beam’s cross-section, the ability to tune the wavelength, and the pulse characteristics. Cost is also a factor. That gives an envelope which can be pushed out at many points to meet the needs of an application, and no company can keep ahead on all possible fronts through R&D. However, IPG has expanded by expanding fiber, so it started by being less diverse than laser makers using various laser technologies to address a range of applications.

Seam stepper

I’m optimistic about IPG, especially in high power, auto, and for the seam stepper. From the transcript:

    “Valentin Gapontsev

    Seam stepper – includes fiber laser very typical from total (inaudible) of what fiber laser inside of the seam steppers controller but also its the various special optical (inaudible) we sold largest capacity with major – if you press time where that could be used without any protection ISF installed once it open the reserve of their major problem with laser they now they have to put the laser automotive in large cabins very expensive each such cabin cost up to $1 million it’s all create a lot of problem addition of what, with our seam stepper first time which qualified to use without any protection against so far. It’s enormous step required but instead by would first introduce such technology in the market.”

If that means a seam-stepper can be used without having to spend $1 million on a light-tight safety cell, it’s a huge plus, especially when competing laser-systems companies can’t avoid the cost. This explains how it’s done (or was done in 2010), but doesn’t estimate the saving:

Fiber laser spot welding” Dr Klaus Krastel, 03/03/2010. industrial-lasers.com

The principle of requiring physical contact before a dangerous force is released, is fairly basic. I haven’t found IPG’s Laser Seam Stepper patents (pending in 2012), but the narrower the claims are, the easier they are to defend, but the easier they are to get around.

A quick look at Trumpf’s lasers seems to bear out that their welding systems which don’t require a safety-cell (like this) are for smaller pieces, or for pieces with rotational symmetry which the TruLaser Cell Series 1000 is used for.

Rotational symmetry illustration

Lathes use rotational symmetry but they can cut screws (and quite complex pieces if fitted with indexing plates, elliptical chucks etc), so I can’t assume that Trumpf’s TruLaser Cell Series 1000 is restricted to absolute rotational symmetry as shown in my illustration.

Heavier welding needs the safety cell (e.g. the TruLaser Robot 5020), but that’s not a thorough check and is only for Trumpf.

IPG need to keep ahead in systems. Trumpf have a lot of experience of supplying machine tools and lasers to industry. They know their customers’ needs and have designed sophisticated laser systems. Trumpf will be looking to their systems expertise when their lasers can’t compete with IPG’s. IPG’s sales of integrated laser systems are relatively small, and are included in an assortment of low-selling lasers and components which added up to $6.2 million revenue in the quarter, only 3.7% of total revenue.

Additive manufacturing

IPG have been supplying lasers for additive manufacturing for over a decade. There’s been good growth in plastics, but they’re expecting more growth in metals, with customers ordering higher-powered lasers to make larger parts, and 10 kilowatts is probably the highest power IPG laser used so far in this area. The applications are more general than those usually considered to be 3D printing, for instance in cladding, using high value-added systems.

‘Cladding’ in general includes construction applications, and optical fiber including active fibers in lasers, but metalworking is probably more relevant here. The wikipedia page for Cladding (metalworking) includes laser cladding, and the principles are explained but without examples. Metal wire or powder is melted on to a metal surface, often to apply or repair a wear-resistant layer. This rather specialist blog has diagrams and a picture of laser cladding in action, with a 4kW direct diode laser from Coherent. It looks like soldering, melting a wire onto a metal surface, but using rotational symmetry. IPG’s ‘high value added’ systems probably allow more complex shapes to be cladded, while minimizing the human intervention required. IPG’s FAQs page says that a single fiber laser can be focused to a 100 micron spot for cutting, to 200 micron for welding, or to 400 micron or larger for annealing or cladding.

I couldn’t resist this link describing how laser cladding is used to restore marine turbochargers, because of the picture of two warships at the top with a shopping basket underneath:

Laser cladding pays dividends for TruMarine” Shiprepair & Conversion Technology: 3rd Quarter 2013.

Electromagnetic spectrum

When you read something like 140 nm in an article it’s hard to know what it means if you aren’t a scientist. Wikipedia’s Electromagnetic spectrum page has a chart and tables on the right, but there’s more variation between tables than I’d expect from the fuzzy definitions of spectrum bands. For example, compare rp-photonics.com‘s numbers for UV to Wikipedia’s. Tables showing powers of ten usually need to be converted using the table on Wikipedia’s ‘Unit of length‘ page.

I’ve compiled lists from powers-of-ten charts, but I might have picked a bad chart (don’t trust any with centimeters in the heading) or converted powers-of-ten wrongly. The waveband in bold is between the wavelengths on either side. There’s no subdivision, for instance the various radio bands. In practice, bands merge gradually like colors in a rainbow, with no agreed boundaries.

Increasing wavelength
< 10 pc Gamma rays 10 pc X-Rays 10 nm Ultraviolet 400 nm Visible 700 nm Infrared 100 µ Microwave 10 cm Radio >10 cm
Decreasing wavelength
>10 cm Radio 10 cm Microwave 100 µ Infrared 700 nm Visible 400 nm Ultraviolet 10 nm X-Rays 10 pc Gamma rays < 10 pc

1 m (meter) = 100 cm = 1,000 mm = 1,000,000 µ = 1,000,000,000 nm = 1,000,000,000,000 pc
1 cm (centimeter) = 10 mm
1 mm (millimeter) = 1000 µ
1 µ (micrometer) = 1000 nm
1 nm (nanometer) = 1000 pc
1 pc (picometer) = 1000 fm (femtometer)
1 A (Angstrom) = 100 pc


I left a few of the pages linked to below, open in Firefox overnight, and had to restart due to unresponsive scripts. Free solutions include not having too much open at the same time, getting an ad-blocker, and disabling browser add-ons. The worst offenders are probably big corporations who think CPU hogs improve their image. Small organizations using Google ads might be less of a problem.

The possible threat to fiber

There’s still a risk that universities and research institutions will produce an advance in beam-combining that threatens fiber technology. Heat dissipation at high power might be a problem, although I’m not qualified to be sure about that. I’m no longer concerned that IPG are too fixated on fiber to consider alternatives, as they sell fiber-less diode lasers and acquired UV excimer and diode pumped solid state technology (see under ‘UV lasers’, above).

Small innovators TeraDiode and DirectPhotonics have not reported much progress.

According to:

PHOTONIC FRONTIERS: DIRECT LASER DIODES: Making direct laser diodes shine more brightly” By Jeff Hecht, Contributing Editor, Photonic Frontiers, 03/01/2013, laserfocusworld.com

DirectPhotonics managed wall-plug efficiency above 30% in a demo. (Also, laser diodes can reach over 70% wall-plug efficiency in the lab.)

TeraDiode are reported to have had a prototype operating at 46% wall plug efficiency (but without specifying the power), in the June 2012 Laser Focus World. The pdf “Beam combining cranks up the power” by JEFF HECHT contributing editor, is on TeraDiode’s site. If you want a pdf about a TeraDiode laser, they want your phone number and company. IMO they’d be more keen to brag if they had an actual product with truly disruptive specs. You can test that via their ‘Products’ tab.

IPG’s highest wall-plug efficiency has been at least 33% since February 2013 (“Ytterbium Fiber Lasers provide 33% wall plug efficiency.“). I don’t mean to imply that wall-plug efficiency is the only important factor, but TeraDiode in particular stressed the efficiency of their lasers. Also, the efficiency considered to be high depends on other items in the specification, and IPG can’t be expected to achieve 33% wall plug efficiency on all their lasers. A table on IPG’s FAQs page lists a ‘Fiber Laser Ytterbium (Yb)’ at 30%+ wall plug efficiency beating disk by 5% to 10% and lamp pumped YAG by 28%, with diode pumped YAG and CO2 in between.

This readable pdf about high-power laser diodes is by electrooptics.com but available from coherent.com: “Stacking for success“. I can’t download it but can read it with Firefox.

IPG make Erbium Amplifiers for scientific research, with beam combining listed in the 2012 10-K as the principal application.

One defense against new laser technology is to produce excellent and innovative laser-systems.

Economies of scale in laser diode production

Shaanxi Shinhom Enterprise Co. Ltd claim to sell 300 million laser diodes units yearly. They’re listed second on globalsources.com. If economies of scale tail off before 300 million per year, then IPG’s economies of scale in laser diodes are not so special, although they keep the advantages of vertical integration, and there’s the complication that IPG only use single-emitters. Shaanxi are highly rated, promise a quick response and show two ISO numbers, but I don’t know how trustworthy globalsources.com are. Sony, Sharp, Taiwan’s Union Optronics and Osram (Germany) all produce laser diodes. Without some hard facts I can’t be sure that IPG have better economies of scale than other laser diode producers. There are other benefits from vertical integration, and so long as IPG are well managed, they should be in a better position to coordinate the development of a laser and its required components to target an application, than separate companies.

Trumpf Photonics supply laser diodes and other components for Trumpf’s lasers. I don’t want to infringe copyright, so find what’s between ‘largest’ and ‘worldwide’ in the claim in “Trumpf Photonics’ 10th anniversary shows full-steam-ahead outlook” By John Wallace, Senior Editor, 09/29/2012, laserfocusworld.com. The piece mentions ‘pump diodes’ and ‘bar’ but not ‘single emitters’.

On their FAQ page, IPG claim to make more single emitter multi-mode diodes than anyone else. The cost of IPG’s laser diodes has fallen from about $30 per watt in 2004, to under $2 in 2012 (as shown in the Needham presentation). This fall has enabled the price of fiber lasers to come down. From the 2012 10-K:

    “Over the last several years, however, our semiconductor diodes have become more affordable and reliable due, in part, to substantial advancements in semiconductor diode technology and increased production volumes. As a result, the average cost per watt of output power has decreased dramatically over the last decade. Because of these improvements, our fiber lasers can now effectively compete with conventional lasers over a wide range of output powers and applications.”

In the “Stacking for success“ article linked to above, a French laser company’s diodes director claims to sell multi-bar diodes at $1 per watt. There’s currently a 100W 808nm QCW High Power Laser Diode Bar on aliexpress.com for US $150, which is $1.50 per watt. The offers are short lived but it’s not too hard to find the cheapest.

IPG only use ‘single emitter’ laser diodes, instead of laser diodes grouped into diode bars or stacks. See IPG’s technology page for their claims about superior quality and lifespans. There’s more on their “Frequently Asked Questions about Fiber Lasers” under “9. Why do other manufactures quote lower life times on their diode bars?”. In particular, laser diode bars require cooling with pure water passing through gold coated micro-channels in copper, at high pressure. Also, the lifetimes claimed for laser diode bars are measured under constant and undemanding conditions in labs, which are different to real working environments.

The price fall benefits many kinds of laser, including direct diode, diode-pumped thin disk laser, the variety of diode pumped solid state lasers where the gain medium is in a rod, and fiber variants such as hollow core and photonic crystal fiber. Chemical (gas consuming), gas discharge, excimer, free-electron, and flashlamp-pumped lasers don’t use laser diodes.

As the cost of laser diodes falls, it becomes less relevant, and the cost of other components becomes more important. I couldn’t find much evidence for a long term fall in other laser production costs. Cost reductions through integration and scale are reported for the industry, but those factors can’t sustain a long term trend.


Vertical Cavity Surface-Emitting Lasers (VCSELs) are semiconductor-based and are also called VCSEL diodes. They can be produced in two-dimensional arrays on four-inch wafers with little wavelength variation, and according to Wikipedia they “potentially could be much cheaper to manufacture”.

Princeton Optronics produce VCSELs, and there’s a long and fairly technical article on princetonoptronics.com. They list eleven advantages of their VCSELs over conventional edge-emitting laser diodes (which can be single emitters or grouped in a diode bar), and the list is more readable. The advantages claimed include 140% more energy per unit area (with more to come), and the falling cost for high volumes.

Vertical Cavity Surface-emitting Lasers” (rp-photonics.com) includes the point that a VCSEL can only have a limited area without a technical problem that boils down to lasing in the wrong direction, but that doesn’t stop powerful arrays of VCSELs.

High Peak Power VCSELs in Short Range LIDAR Applications” (pdf) 7 May 2013, Journal of Undergraduate Research in Physics, Neil E. Newman, Duncan C. Spaulding, Graham Allen, Mohamed A. Diagne
(if the link doesn’t work try on http://www.jurp.org/)
The article will be too technical for many readers. LIDAR or ‘LIght Detection And Ranging’ is used by the military for range-finding. The authors looked into using LIDAR to detect fast-moving projectiles such as shrapnel at close range, and trigger countermeasures. While it seems like a “Further investigation is necessary” sort of paper, it shows interest in VCSELs for the application. The authors mention efforts to use wavelength beam combining with large arrays of VCSELs.

I couldn’t find any evidence of IPG using VCSELs, and the statement that they only use single emitter laser diodes implies that they don’t use VCSEL diodes.

If Princeton Optronics’s list of advantages for VCSELs is correct, then single VCSELs might replace ‘single emitter’ laser diodes, but their power could be limited by the area constraint claimed on rp-photonics.com. Princeton’s list also suggests to me that VCSEL arrays might replace laser diode bars, but only a few of the advantages in the list are extended to VCSEL arrays, although I’m not sure that was always intended. Arrays share the lower wavelength spread and lower temperature dependence, but not the operating temperature and reliability advantages. By ‘replace’ I don’t mean entirely, as laser diodes haven’t entirely replaced flash-lamps as optical pumps. It’s a bigger step for VCSEL arrays to replace single emitters. Princeton Optronics say they are still improving VCSELs, and ask readers to check for the latest specs. I can’t predict how successful they’ll be.

Advances and applications


‘Optical supercontinuum’ means white light with a smooth spectrum, when the application is illumination for human vision. There are also supercontinuum applications where it’s better to have the spectrum extended into ultraviolet, and in biology many molecules absorb UV light and fluoresce. A supercontinuum can be generated by shining a laser beam through a glass fiber with tiny holes, with no need for dopants (small amounts of other substances in the glass). Tweaking the holes changes the output, for instance to get UV. The cost is coming down.

This is fairly long and technical: “SUPERCONTINUUM SOURCES: An even brighter future awaits supercontinuum fiber lasers” ADAM DEVINE and ANATOLY GRUDININ, 06/10/2013, laserfocusworld.com

New Form of Optical Fibre for Laser Light Conversion” University of Bath. Probably a few years old.

Supercontinuum fibers use ‘non-linearities’ which spread the spectrum more, as more energy is put in, and the advance from Bath Uni means that less power has to be put in to get white light out. The technology is not reversible, so you can’t shine a spectrum of light into one end and get a single color out the other. The ‘fibers with holes’ have the more impressive term, “photonic crystal fibers” or PCF.

NKT Photonics puts PCF and hollow core fibers into cool-looking lasers with applications in bio-imaging, metrology/sorting, sensing (LIDAR again), and material processing. They claim many benefits for “using Photonic Crystal Fibers (PCF) for high power fiber lasers” in material processing, and for hollow core fibers in gyroscopes, where stability is essential. If their graphics are accurate, there’s more hollow than glass in their hollow core fibers, with advantages when the beam transits between fiber and air, but it might need more optics if anyone wanted to get the beam into a solid delivery fiber. NKT also supply parts so you can make your own custom-laser.

The relationships between NKT Photonics, University of Bath, the Technical University of Denmark and GLOphotonics are described in the press release pdf “NKT Photonics gives license to PCF technology“. NKT Photonics is owned by over-diverse NKT Holding A/S.

The CEO of GLOphotonics uses the ‘D’ word in “… bringing a new highly disruptive product into the photonics market.”

Leica Microsystems’ “White Light Laser – The Ultimate Source for Confocal Microscopy” also uses a photonic crystal fiber.

Googling “solid state white light laser” brings up plenty of references to research papers but nothing more concrete (as far as I checked). It’s sometimes claimed that there aren’t many technologies for producing a broad spectrum of light. Incandescence (as in old light bulbs) is well known. Newer methods include the use of metamaterials (including photonic fibers) and quantum dots, which I’ve seen described as like artificial atoms so they might belong under metamaterials. Light that appears white can be produced by mixing colors, and a graph of the spectrum would have three or more peaks, or spikes if the colors are pure. Although efficient white LEDs using quantum dots can be made, and quantum dots have applications with laser diodes and VCSEL diodes, I haven’t found any white laser diodes or white VCSELs.

There’s a way of combining beams which is like the reverse of a prism splitting white light into colors, using a diode bar where neighboring diodes emit beams with slightly differently colors, which are combined by the optics into a white or broad-spectrum beam. The smoothness of the spectrum depends on several factors including the number of diodes in the bar and the characteristics of each diode.

Lasers with photonic crystal fibers add to the ways of producing a white laser beam with a smooth spectrum.

One way to mimic white light might be to repeatedly sweep the frequency through the optical spectrum from red to violet. I don’t know if that’s feasible, but frequency-sweeping or ‘chirping’ lasers exist, capable of very high sweep rates. This rather dry Abstract on caltech.edu claims a sweep rate up to 10^16 Hz/sec, at around 1500 nm in the microwave band, with applications in range-finding and 3D modeling. Four VCSELs were stitched together with each handling a quarter of the sweep. I’ll speculate that if digital lasers (see below) live up to their promise, the technique could be used for frequency-sweeping, but I can’t say how that would compare to existing methods.

The HOFGLAS paper (see below) states that hollow-core photonic crystal fibers can produce a frequency comb, a series of spikes in the spectrum which is an alternative to a continuum or one or a few spikes. I expect that a frequency comb could be spread into a continuum with much less spreading, but if that involved using two photonic crystal fibers (in series) there might be no advantage to the method, unless the spreading could be varied.


Hollow-core Optical Fiber Gas Lasers (HOFGLAS): a review” (pdf with a massive URL, 14 authors and 14 pages)

The relatively readable paper describes the advantages and constraints of gas lasers and fiber lasers, which motivate a combination with the advantages of both. Fiber lasers are compact, rugged and efficient, but unwanted scattering and thermal lensing increases as more power is put into the fiber, until the fiber is damaged. Apparently in 2012 the limit was a little over 10 kW for a single fiber, which is only 5% of what an electrically-driven gas laser could produce. (Chemical lasers can output more, but consume gas.) HOFGLAS prototypes have been demonstrated, but only for mid and near infra red. Changes such as using chalcogenide-glass will be needed to extend the spectrum of possible wavelengths, but the wavelength of gas lasers ranges from IR to UV, which is given as an advantage that HOFGLAS could embody. Wikipedia’s Laser page (already linked to) has a diagram showing wavelengths of gas lasers.

Hopefully, the photonic crystal fiber expertise of the Mobius acquisition should help IPG to keep up in this area. HOFGLAS and photonic crystal fibers might not have gone unnoticed by competitors. I can’t be completely certain of the importance of this to IPG without knowing how easily they can combine the output of many fibers, but the power-per-fiber is likely to be more important when beam quality matters more, or coherence is needed with a power which is high for a single fiber.


There are a few paragraphs about synthetic diamond maker Element Six in the “Stacking for success” pdf linked to above. Powerful lasers need to have heat conducted away from hot-spots, and diamond is an excellent material for that, with good optical qualities and able to withstand pressure. Diamond has also been used as the gain medium, with a diamond crystal doing what a fiber does in a fiber laser. (Element Six are ultimately owned by mining company Anglo American, in my opinion not an attractive investment.)

Diamond cooling has been advocated for laser diodes, to prolong their apparently short lives. I haven’t found much actual diamond cooling except for a Yb:YAG microchip laser, and diamond cooling of a diamond crystal. This suggests it’s used in small quantities to conduct heat from small hot spots, probably due to the expense. The expense is supposed to be coming down (and this is synthetic diamond), but I wasn’t expecting thin disk lasers to be stuck to slabs of diamond anytime soon.

A Trumpf thin-disk laser has been glued to a diamond heatsink: “Thin-disk laser for exotic atom spectroscopy” Eth Institute for particle physics.
The one-page pdf is mostly pictures and diagrams. The system also features adaptive mirrors. Google’s search result included “mounted on a 1.4-mm thick diamond heatsink, soldered on a back-cooled copper mount”.

A key advantage of fiber lasers is that it’s easier to keep them cool, and it’s possible that diamond could help to bridge some of the gap for other laser types, although diamond cooled laser diodes would help fiber lasers (and some other kinds), if cheap enough.

Chemical vapor deposition (CVD) can be used to apply a very thin diamond coating to laser media for improved heat dissipation, but I haven’t found any interesting detail for that, or for how diamond coatings can be applied using lasers, or for how pulsed laser deposition applies nano-thin diamond-like coatings. However it looks like lasers can stay cool with synthetic diamond from very thin to at least 1.4 millimeters, or with various diamond alternatives including a nanoscale coating.

See also “Diamonds are a Laser’s Best Friend” under ‘Raman lasers’.

Boron arsenide: potential competitor to diamond as best conductive cooling material” By John Wallace, Senior Editor, 07/08/2013.

Raman lasers

In a Raman laser (Wikipedia), when photons from an optical pump are absorbed they are emitted almost immediately as lower-frequency photons. Like free electron lasers (FELs) and excimer lasers, laser light is not absorbed by the gain medium. Raman lasers are more tunable than most types (but not FELs), with the wavelength depending on the wavelength of the pump-light that powers them, but are low-power, only managing tens of Watts.

Diamonds are a Laser’s Best Friend” University of Strathclyde

Strathclyde Uni claim that using diamond as the gain medium in a Raman laser allows the wavelength to be shifted. They’ve demonstrated such a laser (converted from a semiconductor disk laser) with continuous wave at 5 Watts.

There’s more about the Strathclyde laser here, describing the advantages of using diamond, including a smaller size. Some of the new colors are suitable for some retinal surgery, and the continuous wave is better for eye-safety than pulses. The researchers worked with synthetic diamond maker Element Six.

World’s first ‘tunable’ diamond-based Raman laser” By Edd Gent, 8 August 2013. eandt.theiet.org

A Raman laser using a crystal diamond on a diamond heat-spreader, reached 24.5 Watts in 2011 (which is high for a Raman laser). This paper is free but very technical:
High average power diamond Raman laser” 2011

Raman fiber lasers lase between fiber Bragg gratings in the fiber. In a Bragg grating, light passes through transparent layers which alternate between a high and a low refractive index, a bit like a stack of alternating sheets of glass and clear plastic. This is speculative and sounds like an ‘arm waving’ argument: if there’s a way to change the refractive index along a section of active fiber, it might be possible to ‘write’ a fiber Bragg grating and change it’s properties, possibly with some of the benefits claimed for the digital laser.

See also ‘Raman spectroscopy, lipstick and lasers’.

IPG sell Raman lasers and make their own Bragg gratings: “Telecom Raman Lasers and Amplifiers” ipgphotonics.com, enabling communication over fiber optics beyond 350 km without amplifiers or repeaters, with modules rated at 0.5 Watts to 10 Watts.

Photonic integrated circuits (PICs)

This is about integrating small optical components in an optical chip, in the same general way that transistors are integrated into electronic silicon chips, but currently with only hundreds of optical components on a chip. There won’t be much direct impact on IPG’s main business of materials processing.

Photonic integrated circuit (wikipedia)

Asia emerging as photonic integrated circuit power says report” Lightwave Staff, July 15, 2013, lightwaveonline.com

Infinera’s photonic integrated circuit named Best Optical Component Product Beyond 100G” (Marketwired – Jun 20, 2013), finance.yahoo.com

Intel is using conventional CMOS technology to put lasers on a transceiver chip:

Intel’s Justin Rattner on New Laser Chip Business” By Tom Simonite, July 2, 2013. technologyreview.com

Single-photon lasers

This depends on a cloud of atoms cooled to near absolute zero…

Single-photon transmitter could enable new quantum devices” David L. Chandler, July 25, 2012, MIT News Office.

… but this is solid state, using quantum dots in a Schottky diode, and there’s no mention of temperature.

Laser-like photons signal major step towards quantum ‘Internet’” Mar 19 2013, University of Cambridge.

(The only connection I can find between IPG and quantum dots is an old patent: “Optical fibre with quantum dots CA 2199506 C” filed in 1995 and published in 2001.)

Digital lasers

The man behind SA’s laser breakthrough” 22 September 2013. techcentral.co.za
Mostly about the modest inventor, but with a simple description of the invention. The comments under the article are political.

This page plays video without waiting to be asked, but right now there isn’t much choice for articles with any technical detail:

SA claims laser breakthrough” By Nicola Mawson, ITWeb deputy news editor. Johannesburg, 17 Sep 2013

A laser’s mirror is replaced by a digital image, which can be changed. There’ll be no pictures of cats, instead there’ll be optical patterns to change the beam’s properties. Lasers generally involve reflection, but not always what you’d call a mirror, and an active fiber usually reflects from the ends of the fiber (light bounces from one end to the other until it escapes), unless reflection is between Bragg gratings inside the fiber. I’d have thought the technique would be easiest for reflection between relatively normal mirrors, with gas-mirror-gas reflection rather than glass-boundary-glass or any internal reflection in a solid, but from this Abstract on nature.com:

“The phase and amplitude of the holographic mirror may be controlled simply by writing a computer-generated hologram in the form of a grey-scale image to the device, for on-demand laser modes. We show that we can digitally control the laser modes with ease, and demonstrate real-time switching between spatial modes in an otherwise standard solid-state laser resonator.”

‘Solid state’ means solid but not semiconductor or fiber, and usually means the gain medium is in rod form.

Whatever benefits digital lasers bring, they’ll apply to the lasers where the technique can be employed. Even if it can be applied to fiber, a powerful gas laser only needs one or two ‘digital mirrors’, but a laser combining several fibers to get the same power would need that for each fiber.

It’s hard to rate the story with certainty until informed reaction to the paper on nature.com emerges.


The next link could be slow to load as the video buffers up. The laser is a FEL or free-electron laser, described near the end of the text.

The future is now: Navy to deploy lasers on ships in 2014” By Justin Fishel, Published April 08, 2013, FoxNews.com.

FELs “generate coherent, high power radiation, that is widely tunable, currently ranging in wavelength from microwaves, … to soft X-rays.” (Wikipedia’s Laser page)

In most lasers, an atom or molecule falls back to a low energy state when it emits a photon, which makes it ready to absorb another photon. In an optically pumped laser, you need photons from the optical pump to be absorbed, but you don’t want photons emitted in the gain medium to be absorbed (at least not when they’re going in the right direction). You want photons with the right wavelength and direction to hit an atom in a high energy state to stimulate more emission, or come out the business end, especially if you want a high energy pulse. A FEL avoids the problem of unwanted absorption because the lasing happens in free electrons, instead of in normal atomic material like gas, liquids or solids. From Wikipedia: “FELs use a relativistic electron beam as the lasing medium…” and “…it appears that the operation of this rather exotic device can be explained without reference to quantum mechanics.”

$40 Million Laser Weapon System Installed On A Destroyer” By: David Russell Schilling, June 23rd, 2013. industrytap.com

Boeing solid-state laser weapon system outshines expectations” By David Szondy, August 18, 2013.
There isn’t much about efficiency except for 30% better, or about weight.

The science of beam weapons” By Graham Templeton on April 18, 2013, extremetech.com
This is a simple description of the limits, such as how a Star Trek style phaser would need to store a vast amount of energy in a small volume.

Kickstarter sci-fi book to pay for laser weapon research Technology” by Philippa Warr, 16 January 13, wired.co.uk/news.

Smaller particle accelerators

Laser research lights way to mini particle accelerators” By Stephen Harris, 15 April 2013, theengineer.co.uk

The Petite Particle Accelerator: A Proton Gun For Killing Tumors” Spencer Woodman, 3/26/12, gizmodo.com.
From the comments, the tech might be well known, it’s the market that’s uncertain.

Turbulence warning

DELICAT LIDAR system alerts pilots to air turbulence ahead” Posted by Gail Overton, Senior Editor, 08/07/2013.

The density of the air fluctuates more than usual in turbulence. This can be spotted by pointing the laser forwards and measuring the radiation scattered back from air molecules. DELICAT is a European Union project with many partners, and one of them is likely to have supplied the laser, possibly Laser Diagnostic Instruments. IPG’s AQS Ho:YAG Series of lasers can be used for LIDAR, as well as materials processing and medical applications.

Laser headlamp

Light by Laser Next Headlamp Trend, Experts Say” by William Diem | WardsAuto, Oct. 17, 2012

One idea is to shine laser diodes onto a phosphor screen to get a bright white light. VCSELs might be the solution to problems about the life span and temperature sensitivity of laser diodes. Princeton Optronics’ list illumination as an application for their VCSELs.

The patent “Vehicle headlight US 7150552 B2” (google.com/patents) is for using diode arrays including VCSELs for a vehicle headlight, capable of various adjustments without any mechanically moving parts. Including VCSELs as a possible source in a patent might not mean a lot, as there’s no incentive to leave them out if they aren’t likely to be practical.

A scanning beam could enable sophisticated applications such as illuminating pedestrians without shining the laser in their faces. IMO there could be eye-safety issues.

The ‘phosphor screen’ idea has also been thought of for: “Outdoor Street Light Fixture with Novel Laser Diode Light Source

Medical devices

There’s nothing stunning in this but it’s a good round-up.

Lasers Help Shrink and Sharpen Medical Devices” Hank Hogan, Contributing Editor, hank.hogan@photonics.com, July 2013

Laser-propelled red blood cells

Laser propulsion and deformation-monitoring of cells in optofluidic photonic crystal fibre” 21.06.2013

A short technical paragraph about using lasers to shift blood cells, using less power than previously. The method should help scientists to study the biomechanical properties of single cells, which can indicate how healthy the cell is.

Raman spectroscopy, lipstick and lasers

New forensic technique for analysing lipstick traces” University of Kent

A laser is shone at a small sample inside a transparent evidence-bag. Light reflected back at the laser’s frequency doesn’t tell you much about the sample, but the other light reveals the energy of molecular vibrations, from which the composition of the sample can be deduced. A low power pulsed laser is used, with enough energy in the pulses to do the job, but too weak to damage the sample. The lack of damage, sensitivity and ability to keep the sample in the bag allows ‘continuity of evidence’, and the test can be repeated later if there is any doubt about the result. Keeping the evidence intact means that other tests can be performed, including tests invented later.

Top end Raman spectrometers like the one used here cost hundreds of thousands of dollars, and I’d guess the laser needs a pure beam that can be focused (if necessary) onto a tiny spot, but isn’t expensive compared to the rest of the kit. Here’s a technical thread about making your own, for a few hundred or a few thousand dollars: “Cheap, Low-Resolution, Raman Spectroscopy” 18-2-2013. sciencemadness.org. There’s an illuminating sub-topic about how the power of laser pointers is often under-rated to clear customs, with a contributor slightly alarmed at having used one in court (I won’t spoil ‘Really gives a literal dimension to …’).

Laser pointer or lightsaber

Your Laser Pointer Is Probably Illegal” By Jason Bittel, March 22 2013. slate.com/blogs

The story includes a link about twelve cases of laser blindness, when lasers meant to be pointed at the sky were used indoors, at a rave in Moscow in 2008.

Printable lasers

Researchers develop printable lasers” Sep 19 2012, University of Cambridge.
Put the magic ink in the inkjet, and print your lasers.

Digging out fossils

This asks if you want to download a pdf, but you can say no and it stays in the browser.

An investigation of Laser Induced Breakdown Spectroscopy for use as a control in the laser removal of rock from fossils found at the Malapa hominin site, South Africa” Roberts, du Plessis, Steyn, Botha, Pityana, Berger, 2012. academia.edu

I heard something on the radio about ‘dino-drills’ to dig up fossils without too much damage, and found the link above for using lasers. It’s in the academic style so not much enthusiasm, but has ‘great promise’ in the conclusion.



The image is from NASA and is in the public domain, according to Wikipedia. Watch it fly on YouTube: “Rocketships Laser Propelled Light Craft” (it starts with a small version).

It’s probably from a few years ago. According to Wikipedia, “In 2000, a new flight record was set with a flight lasting 10.5 seconds and reaching 72 meters (236 feet).” The problem with conventional rockets is that a huge weight of fuel is lifted a long way. That’s avoided if fuel can be burned on the ground to power a laser instead. These days scientists just seem to write papers about laser propelled vehicles.

Death watch

Scientists invent skin-prodding laser device to tell you when you’ll die. Thanks, science.” By Natt Garun — August 12, 2013, digitaltrends.com

What ‘death watch’ makers need is a positive way to fit into the ‘Self tracking’ trend. I haven’t found much about lasers on quantifiedself.com, but there’s this:

Measuring Vital Signs From 40 Feet Away” Posted on June 1, 2009 by Alexandra Carmichael

Yes, the laser ‘tricorder’.

Tons more

Googling ‘laser levitation’ currently gets a story about levitating tiny diamonds, many times over.
I like the result of Google truncating a title: “Nanodiamonds Levitated With Laser Light, Scientists Held Trapped …”.

Googling ‘physorg laser’ for the past year, currently gets “New high-tech laser method allows DNA to be inserted ‘gently’ into living cells”, “Interplanetary precision laser could reach to Mars and beyond”, “Laser communications set for Moon mission – Science News” and many more. Articles on physorg aren’t all high quality and don’t always live up to the headings. Their Optics and Photonics page lists stories, currently ranging from gravity waves to claimed ‘five-dimensional’ data storage (the normal three plus size and orientation).

Googling ‘Laser breakthrough’ gets “Hope for blindness cure…”, “CHILLERS AND COOLERS: Breakthrough of optical refrigeration …” (for semiconductors), a skin-resurfacing treatment, “Laser 3D Printing Breakthrough in China…”, even “Laser Breakthrough in the Treatment of Snoring” but that one’s from 1993.

The big picture

IPG can’t stay ahead in every area of laser research. Having plenty of new applications exploited by other laser-makers means that IPG can choose which area to enter next. Some technologies are irrelevant or only relevant as possible opportunities in the future, and a single photon laser will pose no threat to a welding system. There’s a risk that a specialized technology will balloon, allowing a small company to expand in the same way that IPG grew by developing it’s fiber technology, protected by industry-wide skepticism about fiber lasers and how far IPG could expand the area of application. To the extent that IPG’s competitors were less vertically integrated, IPG should be a harder nut for an upstart to crack.

Some areas of tech have had leadership changes, with Nokia, Yahoo and MySpace falling behind competitors. The laser industry seems more stable, but it isn’t impossible for a breakthrough to build over years into a leadership change. It’s unlikely that a few adjustable ‘superlasers’ will replace the current product diversity (unless I’m under-estimating the digital laser), so any falls should be gradual or relative, with time to adapt where management are capable of it. (See under ‘Specification diversity’ above.) While the term ‘disruptive’ is overused in the industry, genuine “Disruptive innovation” (Wikipedia) can’t be entirely ruled out. Disruptive innovation typically happens at the low end, and there’s some reassurance in IPG launching lasers to compete with low cost lamp pumped YAG lasers. It’s also good that while IPG have high margins, they are not margin-fixated. The diversification into new applications reduces any vulnerability, and IPG’s strong balance sheet provides time and possibly the means to respond to any kind of innovation by competitors. Vertical integration could make fundamental change harder. (BTW, some fans of Disruptive innovation theory might accuse me of subscribing to the ‘mudslide hypothesis’ for saying ‘falling behind competitors’.)

There isn’t much on IPG’s technology page, they focus on products and applications, and I found they were involved with photonic crystal fibers (through the Mobius acquisition) from optics.org, not from IPG. They could have technology I’ve mentioned, without my knowledge.

Added September 28 2013 (down to ‘Valuation’)

Solar Pumped Laser

Solar Pumped Laser” May 17, 2013. solarharyana.com

Not much use where the grid is available, but the ultimate patio curiosity, going by the pic. Apart from nanopowders, suggested applications are in space, including propulsion, powering satellites, and defending against missiles. Back to the pic, the disk is likely to be a Fresnel lens (Wikipedia), which is flat. This focuses sunshine into a rod of whatever the gain medium is. The other works are probably to track the sun.

Multi-Fresnel lenses pumping approach for improving high-power Nd:YAG solar laser beam quality” Dawei Liang and Joana Almeida. opticsinfobase.org (Abstract)

With the lenses, some calculation, and the crytal or rod suggested, the authors claim to improve brightness by 1300 times. (Brightness is roughly power per unit of area.)

Design of a formation of solar pumped lasers for asteroid deflection” University of Strathclyde

The readable Abstract is about deflecting an asteroid in a dangerous orbit. Solar pumped fiber lasers are shone from a formation of spacecraft onto a spot on the asteroid. The material there vaporises, turning the asteroid into a jet-propelled asteroid.


This thread alerted me to the importance of relationships between competitors and their customers:

this might be huge, and probably will happen” Thread started by Itgmaster, Sep 24, 2013. finance.yahoo.com

Itgmaster’s reply highlights the relationship between JDS Uniphase and a customer, with quotes from JDS Uniphase. A relationship like that, or inertia for any reason, will give IPG an incentive to reduce their prices. If a lower price is enough to overcome inertia, IPG can later increase the price gradually, or get their margin back by reducing costs. I don’t have evidence that IPG operate like that, beyond the quotes about pricing and margins. I see flexible pricing as different from across-the-board permanent price cuts, which would have an obvious effect on margins, and force competitors to react. IPG won’t want to make materials processing a low-profit area, and big rivals with lower margins won’t want it either. Hopefully, competition will be mostly non-price, and lower pricing will be used with caution, otherwise there is a risk of escalation.

The first evidence I found for old margins is from 1993 (possibly a year out): “increased our gross margin to 46%” Rama Rao, Chairman and CEO of EXCEL Technology, Inc. EXCEL’s operations spanned the laser market.

From IPG’s 10-K for 2006: “In the last three years our gross margins have increased from 30.4% in 2004 to 35.2% in 2005 and 44.2% in 2006.” Then it’s 45.0% in 2007 and 46.8% in 2008, showing that gross margins were unaffected by the dip in 2007.

(see ‘Edgar database’ under the ‘Links to free resources’ tab, for getting 10-Ks.)

I hope that tiny sample corresponds to generally high margins in the past, which continue.


TTM and Most recent quarter ratios are from Key Ratios page on IPG’s website.

Projected EPS Q4 2012 to Q3 2013 calculated by:
= (Net income attributable to common shareholders Q4 2012 to Q2 2013 + (Q3 2013 income derived from guidance) ) / 52,385,000 diluted shares

Thanks SA

With thanks to Seeking Alpha (seekingalpha.com) for their policy about quoting from their transcripts, which is stated at the end of the transcript. I’ve quoted 266 words, and there’s plenty worth reading in the transcript that I haven’t quoted.

DISCLAIMER: Your investment is your responsibility. It is your responsibility to check all material facts before making an investment decision. All investments involve different degrees of risk. You should be aware of your risk tolerance level and financial situations at all times. Furthermore, you should read all transaction confirmations, monthly, and year-end statements. Read any and all prospectuses carefully before making any investment decisions. You are free at all times to accept or reject all investment recommendations made by the author of this blog. All Advice on this blog is subject to market risk and may result in the entire loss of the reader’s investment. Please understand that any losses are attributed to market forces beyond the control or prediction of the author. As you know, a recommendation, which you are free to accept or reject, is not a guarantee for the successful performance of an investment.


2 thoughts on “IPG Photonics – growth, cash, markets and technology

  1. From my previous piece about IPG: “June 2012, regained full ownership of Russia-based subsidiary, NTO IRE-Polus, a pioneer in parallel pumping, allowing multiple diodes to be coupled to pump a single fiber.”

    This is a reply to the_stock_warrior, which I submitted twice on Yahoo, but each time it disappeared later. (With small changes.):

    Thanks the_stock_warrior, and thanks for reminding me about Polus. I did a quick search and found a piece on Efiber.net with this:

    “IRE-Polus is the only company employing parallel pumping, as well the incorporation of multimode diodes with single mode fibers. Both employ a patented coupler technology.” (undated)

    My other searching hasn’t proved anything definite. I found Onex Automation and Beamer with lasers for marking, with patented fiber laser parallel pumping, but I don’t know who makes the lasers or owns the patents. For Coherent and Trumpf, nothing I found was definitely fiber laser parallel pumping.

    I think maybe disk lasers could be matched by ‘fiber disk’, where the fiber is wound into a flat disk and the pump diodes shine onto the side of the disk.

    I’m amazed by the tech I find when I’m looking for something else. I just found a research abstract called “Design of a formation of solar pumped lasers for asteroid deflection” with fibre lasers proposed.

  2. I’ve demoted this to a comment because it’s too speculative:

    Laser-cut Fresnel lenses

    Wikipedia says, about Fresnel lenses, “In modern times, computer-controlled milling equipment (CNC) might be used to manufacture more complex lenses.”.

    This patent: “Fresnel lens laser cutting method” google.com/patents, was deemed withdrawn in 2006. The reasons given for why traditional methods are bad, include creating cracks that can spread, and that might still be the case.

    Inexpert speculation: With IPG lasers now cutting glass, and recent expertise at the finer end of materials processing, could IPG could make lasers that cut Fresnel lenses? My guess is ‘No’, I can’t find any actual laser-cut optics, and I’ll probably regret writing this. If the idea isn’t laughable, Fresnel lenses have many applications, from lighthouses to solar cookers. The widest application, vehicle headlamps, is giving way to multifaceted reflectors. If laser-cut Fresnel lenses are possible but very expensive, they might be worth considering for telescopes in space, if current methods really do cause problems.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s