New EYEMAGIC Scanners EMS7000

[norty303]

I guess no-one is willing to talk then, as I understand some testing did happen at a recent US LEM. Read into that what you will.

[andythemechanic]

I read into that, they turned out not to deliver what was promised?
Well I was also waiting for some statement about the performance of this scanners. But in the meantime I bought some CT6210 which probably was a better choice.

Andreas

[Pangolin]

I take issue with this sentence. Your scanner does not exist.

Well that's a fair point. But my comment wasn't to compare our development to Tom's -- only to respond to his comment about making a cheap scanner. Our intention is not to make a cheap scanner, it's to make a great scanner. Hopefully the greatest!

You haven't proven anything. Bill. Not that there aren't valid complaints about some of EyeMagic's numbers - the beam aperture vs diameter issue in particular - but really...

First let me say that my main problem with the EMS-7000 is the advertisement. It is advertised as being literally the fastest in the world, but it isn't. If the advertisement was more modest, then I would never have gotten involved. But to claim that it is the fastest in the world, and to have discussions of how it's better than a Cambridge model 6215, and then finally on top of that, to have a datasheet which has data which is flat out incorrect (the mirror size versus aperture is only one of the areas in which it is incorrect), then yes, that's what I have a problem with. And this isn't about Tom. If *anyone* makes such false claims, they can expect a response...

Would you care to provide, or at least enumerate, these formulas? Can you say in more detail where exactly that 40kpps number came from?

I'll be happy to clarify anything, and respond -- factually -- to any question. No problem.

The K rating of a scanner has more to do with heat inside the scanner than anything else. That's why a Cambridge model 6800 is a 30K scanner period. And that's why all scanners that look like a Cambridge model 6800 are also 30K scanners. Sure -- scanners may be tuned to a higher rate. So doesn't that make them faster? NO!! The reason is because -- to tune them to a higher rate will generate far more heat inside the scanner to the point of failure. The flip side of this is that higher tuning (with the same torque, same resistance, etc.) necessarily requires the size to be reduced -- geometrically.


Bill

[Pangolin]

I read into that, they turned out not to deliver what was promised?
Well I was also waiting for some statement about the performance of this scanners. But in the meantime I bought some CT6210 which probably was a better choice.

Andreas

There are two kinds of 6210. Old, and new ;-) The old ones are essentially the same as a Cambridge model 6800, but with a thicker mirror and much better position sensor. The newer ones (with H on the end of the number) have a different coil making them truly better than a 6800. If we use "heat" as the standard to determine the K rating, the 6800 is a 30K scanner, and newer 6210s are around 40K. Cambridge 6215 are true 60K scanners, and the only true 60K scanners available today (again, using heat as the determining factor).

Bill

[Pangolin]

I guess no-one is willing to talk then, as I understand some testing did happen at a recent US LEM. Read into that what you will.

Much testing was also done at a recent Florida Laser Enthusiasts Meeting. I think a total of 8 scanning systems were tested. I have a document here which details the results and -- as soon as I dig my way out of email land (3500 messages and counting), I'll post some results.

One great thing about the meeting was that I think there were around 20 people present to witness the results (so it's not just the Big Bad Bill Benner doing the talking). Probably a half dozen watched very closely. We projected patterns, made measurements, took pictures, including with digital photo thermometer to see the heat rise, etc.

More later...

[heroic]

(expensive bearings)
They're made in Germany. In the words of the German manufacturer, "they are the Porche, and made with the absolute best material available today".

These bearings are 100% custom made including the size. Because of these factors, a single one of these bearings costs more than the entire manufacturing cost of most scanners today.

I don't doubt they're very good bearings, but more than $600 a set? No.

Could -- theoretically -- but at what cost? Several attempts have been made in the past for digital amps to serve this industry. Probably nobody on this forum is aware of these attempts and developments. They all failed... Nevertheless, Cambridge does have a digital amp, but the cost is far far more, and also there is another cost. Performance in one area is better, but in another area is worst...

What was expensive in the digital domain a couple years ago is not expensive any more. I can get a 200 MIPS 32-bit microcontroller for less than $5 now. I can get one with a vector floating point unit for less than $10. Moore's law is a powerful force in the control industry.

As far as I know (and I really should know with all of the testing I've done on these scanners, and all of the magnet vendors I deal with) Cambridge is using the highest energy product magnets available THAT WILL OPERATE CONTINUOUSLY AT 110 DEGREES C.

I'm going to ignore your appeal to authority and speak to the point that the main market for these magnets is actually motive power; the hybrid car and electric motorcycle markets have really driven the demand for these types of magnets. (The reason for this is that China banned internal combustion for scooters in cities a couple of years ago- fuelling frenzied brushless DC motor development.) The magnets used in CT scanners are samarium cobalt. There are high temperature NdFeB magnets now that are rated for 120 deg C continuous operation. They have a little over half again the energy product of the SmCo magnets.

Hehe. Sure. But with the slotless brushless approach taken by virtually all scanner manufacturers (including the EMS-7000), if you want "lower impedance coils" you're also going to get a lower torque constant.

So don't go slotless. You can run magsims on a modern desktop machine that would have needed a couple hours on a big Cray a few years ago.

Driven by what software? If I buy a Cray computer, capable of calculating a zillion K, what laser software will I run on such a thing?

It would be easy to drive whatever scanner interface I care to use from LSX using an Etherdream; the firmware for Etherdream is open source, so I can modify it to directly drive a digital interface (it already uses SPI or I2S to run its DAC chipset in any case... as, I imagine, does the FB3.) Etherdream supports 120k pps on analogue and can probably do more if the digital interface is optimized.

A different hardware interface does nothing without software to drive it...

So you're basically saying that the big problem is your closed software that does not allow any development on the DAC end, Bill?

671xx high power amps (the ones that drive 6215 a true 60K) are indeed true H-bridge...

Crazy not to, really.

(class H, switched rails, etc.)
Sure, high tech, but at what cost? What's the cost in terms of money and noise?

Samson seem to do a very nice range of servo reference audio amplifiers for less than $300 a piece. There are many audio chipsets that can provide similar service at low cost.

What does ILDA spec have to do with magnet size, magnet field strength, coil impedance, or output drive technique?

I was referring mostly to the antiquated analogue +/-10v signal with no scanner feedback. Positively antediluvian.

It's like our projectors are these advanced Star Trek technology ultra-HD video devices... and we have to drive them with an original IBM VGA card.

[absolom7691]

By the way, there is one thing to remember. Within my last post, I told everyone that I was going on a European tour along with others from Pangolin (Justin and Francesco) for a month. In fact critical members of Pangolin were out of the office for 32 days straight! So we're just getting back into the office and getting settled in. Hopefully that explains the delay a bit...
Bill

This was actually directed at the other members who had purchased these, mainly those who were quick to defend the claims before testing them. I know Pangolin, as well as yourself, have been busy with trade-shows abroad.

[absolom7691]

(so it's not just the Big Bad Bill Benner doing the talking).

Bill, you're not bad.... you're just drawn that way!

[carlos3621]

A pissing contest, or pissing match, is a game in which participants compete to see who can urinate the highest, the farthest, or the most accurately.[1] Although the practice is often associated with adolescent boys, women have been known to play the game, and there are literary depictions of adults competing in it. Since the 1940s the term has been used as a slang idiomatic phrase describing contests that are "futile or purposeless", especially if waged in a "conspicuously aggressive manner".[2] As a metaphor it is used figuratively to characterise ego-driven battling in a pejorative or facetious manner that is often considered vulgar.[3] The image of two people urinating on each other has also been offered as a source of the phrase.[4]

[Jem]

Did you know that the word facetious is one of only two words in the English dictionary with all the vowels in the correct order

On another note... I had a projector with Cambridge CT6215h scanners with high speed amps. I sold it, and I now have a projector fitted with the EMS7000's. Sadly I sold it before I had a chance to do some direct comparisons. However, I'll be absolutely honest and say that from a general projector use point of view I can't really tell that much difference. I know that's not really very scientific, but I just look at pretty patterns on the wall or beams in the air and that's about as technical as it gets with me. I do think the EMS7000's are noisier than the CT's and they aren't quite as sharp at very high speed, but I probably won't ever use them at full tilt as by then they'll be screaming to get out of the box