From: Jim Thompson
Subject: Re: How to increase PLL order?
X-Newsreader: Forte Agent 1.91/32.564
Date: Tue, 10 Dec 2002 00:12:02 GMT
NNTP-Posting-Date: Mon, 09 Dec 2002 19:12:02 EST
Organization: Cox Communications
On Mon, 09 Dec 2002 16:58:04 -0800,
"Christopher R. Carlen" ,
In Newsgroup: sci.electronics.design,
Entitled: "Re: How to increase PLL order?",
Wrote the following:
|Jim Thompson wrote:
|> Chris, Are you trying to get a lock on shaft position or it just
|> speed control you are after?
|> If the latter case I've used a normal PLL (with its own VCO) driving
|> an EC (electronically commutated) motor.
|Yeah, in application #1, I just need speed control. Thus, the PLL that
|I'm using, which is about exactly what is explained in U113 which I can
|understand reasonably well, works very well.
|Application #2 will have two wheels each on their own motors. One will
|turn at 350Hz and has a very narrow slit, about 1 degree of arc for an
|optical shutter with about 8us open time. But 350Hz is too frequent for
|a camera shutter to gate out only one pulse per 100ms (the laser period
|that illuminates the subject), so we divide the frequency down by a
|factor of 7 with an additional wheel, with a larger slit, running 50Hz.
|This was all done mechanically before (in app. #1), but the gears and
|bearings wear out too fast with the fast wheel running at 21000RPM. So
|my plan is to electronically "gear" the motors (app. #2).
|That requires absolute phase lock of the shaft positions, though not
|very stringent phase jitter requirements since I can always detect the
|fast wheel near the top position with my rotation sensor, and trigger
|the lasers from that. So it should be well within my means with basic
|PLL design, needing no exotics to acheive other-worldy phase jitter
|performance. I have pretty much promised and am confident that with
|some further study I can pull this one off.
|Application #3, which may be beyond my means, is to run two wheels, both
|close to 21000RPM or higher but counterrotating, with a difference
|frequency of 10Hz. Absolute phase locking to achieve the slit overlap
|at a repeatable and controlled absolute shaft angle. Considering that
|we need less than 5% jitter timing the laser to when the slits overlap
|in order to avoid excessive amplitude jitter in the signal, then this
|represents a serious challenge! Also, the absolute shaft angle position
|of the slit overlaps must meet that requirement as well. If it is
|possible, this will take us to less than 4us of mechanical shutter
|performance! This will be a test that I will do later after application
|#2 is complete, and there is no loss if I can't make it work. Then
|again, I just might hire one of you experts to help with the hard stuff
|at that point.
|Why do I need a mechanical shutter, considering the fact that there are
|electro-optical shutters that can do sub 100ns or so? The reasons are
|wavelengths, transmission efficiency, contrast ratio, and some others.
|We need mechanical, trust me.
|Thanks for the input.
Aha! I last did a smear camera around 40 years ago in the MIT MHD
labs. But I cheated, the motor (with rotating mirror) ran open-loop
at 20,000 RPM, but I used a sensor to detect when the mirror was
located correctly, then fired the plasma tube ;-)
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice:(480)460-2350 | |
| Jim-T@analog_innovations.com Fax:(480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |
For proper E-mail replies SWAP "-" and "_"
I love to cook with wine. Sometimes I even put it in the food.