Printable View
I'm going to go way out on a limb here and suggest that the time constants and phase delays inherent in the output filtering of such an amp would make the closed-circuit feedback loop response times too large to achieve fast scanner performance.Quote:
Originally Posted by -bart-
But, I could be all wet there, never having measured any of that. ;)
I have wondered the same thing. I am wondering about good H-bridge PWM design. I was able to get 100 hz out of nema 34 ac brushless servo. And i was moving it 30 deg... ( not for laser just testing the servo)Quote:
Originally Posted by -bart-
chad
Voila, here we go:
Attachment 36851Attachment 36852
Looking good! I didn't ask for cost, since I didn't expect n answer. ;-)
Well, color me wrong on my limb!
After intensive googling, looks like Cambridge implemented at least one line of class D amps a few years ago:
"LII PWM Drive: Generates power instead of heat; 90% efficient amplifier"
http://www.camtech.com/index.php?opt...116&Itemid=183
Really good, good, scanner stuff is Class D + DSP..
Look at Lightening IIs spex, 48V rail, optional water cooling, State-Space controller ie DSP.
Next question, how thick is the mirror, how hard to replace etc...
You may be seeing a need for a hotter, stiffer PSU rail with these. I was wrong on the heat guess, he went the other way and went low ohms.
Steve
If you want a better look at the ScannerMax specs docs they're here on the ScannerMax Website:
http://celebritysites.com/beta/scann...des/specs2.jpg
Data Sheet here:
http://celebritysites.com/beta/scannermax/pdf/SM5.pdf
Andy have you got a pair? (of scanners that is!) (Thought I'd better clarify.)
One problem I see (Bill are you listening?) is what they describe as the extended rotor/mirror back support. The image is a bit difficult, but if I read this right then the end of the shaft is cut on its side from an "O" to a "D" to reinforce the back of the mirror. Mechanically, this not a bad idea as this is the point of greatest stress and the added rotational mass near the axis is minimal. However, any change in mirror thickness to accommodate other mirror sizes will force the mirror's center of mass to move in or out and away from the rotational axis = vibration. Only by providing a variety of different shaft cuts or machining the mirror back at the interface to counter this (complex and unlikely) or constraining all mirrors to a uniform thickness can this be avoided. Single thickness mirrors will mean there will be a compromise in ultimate performance and a narrowing in the available choices.
One suggestion, I am working on a modification for the Eye Magic scanners that reduces mirror mass by grinding down the back-outer edges of the mirror and converting the cross section of the mirror, viewed perpendicular to the shaft, from a rectangle to a trapezoid. This reduces the mass where it hurts the most and is unneeded. If the Pangolin engineer does the math, then this relatively straight forward operation will counter the radial shift of mass with a thicker mirror.
Bill is at Photonics so unlikely to see this. Best email him with your suggestion.
The mirror mounting is almost like at the CT's. The rotor has a slit where the mirror is supposed to be glued in. The back side of the mirror just has longer support to eliminate bending resonances of the mirror.
I am not sure at this point if it helps a lot, depends on where these resonances are in frequency. When I am back at home next week I can do the calculation of the resonances with an FEM tool.
Haha That is awesome. Andy thanks for posting this to inform the community. Walking arround on the Photonics event, keep an eye on what,s going on at PL and answer with Info. Thanks for that!!!!:)