Moderators: Dave Moll, Andrew Davie, Steve Anderson
Steve Anderson wrote:A LM317 could be used for motor control, but there are a few problems I see initially...
1) Using the 'control' pin of the '317 as its input the output is 1.25V higher which means unless you have a negative supply and arrange things to suit you cannot shut off the voltage to the motor completely.
2) The '317 is a supply of current, whereas the MOSFET is a drain/sink of current., the topology of the circuit would have to be changed quite a bit. However this would compensate for the major difference that the MOSFET is an inverter, the '317 a follower.
3) There is a current out of the control pin of the '317 that a MOSFET does not have, this would require quite a redesign of the loop filter components.
4) The output voltage would be limited to 1.25V more than the supply of the 4046, 15V + 1.25V, so you couldn't control a 24V (or more) motor.
These are not insurmountable, but quite a task and a complete project in itself.
As for the monostable idea, I don't quite grasp your concept.
Steve A.
gary wrote:I suspect a couple of 555s in BIstable mode would make a a good sync motor control circuit, but the only real advantage over 4 gen purp npn transistors configured in bistable mode would possibly be the inherent pulse signal conditioning they provide.
gary wrote:I've used every possible circuit that you could think of over the years with varying degrees of success. With the bistable approach there is very little that can go wrong and when it does it is relatively easy to debug. The 4046 circuit works well when it works but it can be a right ornery critter to debug when it doesn't, and I, and I get the feeling most others, don't know why.
The only thing that I have modified with the stock standard 4 transistor circuit is I like to condition the pulses such that they (sync and opto) are similar in nature - width and magnitude wise. That seems to make the circuit pretty much fool proof. LOL I use 555s to do that, hence I've been meaning to get around to just using the 555s in bistable mode, just haven't done so yet.
gary wrote:It's a pretty old tried and proven circuit, note, however, that it doesn't give you any sort of frame lock as the 4046 cct tends to. Still, I like it as it is pretty easy to get going and you can easily trace all the signals for debugging.
See newsletter volume 14 number 3 for more info.
harry dalek wrote:Hi Gary
Just reading the newsletter i think i would try it via a 555 so i don't think i would copy the circuit fully apart from the idea block diagram...looks pretty easy but i bet theres a bit of fine tuning to it ...Oh well its a start to try ,thanks for the info .
gary wrote:...See newsletter volume 14 number 3 for more info.
Steve Anderson wrote:gary wrote:...See newsletter volume 14 number 3 for more info.
Although I have perused the back-issue CD somehow I missed this. It's intuitive in it's operation and de-buggering is simple with the exception perhaps of 'R' and 'C' in the diagram. I'm not quite sure of the function of the diode, but presumably it's required.
Instead of discrete trannies any set-reset bistable would function. To be honest I find this more promising than the usual 4046 arrangement, but I haven't tried either as yet. This is still a PLL, make no mistake, the motor/disk/photo-detector comprise the equivalent of a VCO, exactly the same as the more familiar 4046 arrangement.
Now with a 'missing pulse' input and a 'missing pulse' feedback there is a chance that it should achieve frame-lock also. This all hangs on the loop gain, damping and mechanical elements (as ever).
Steve A.
Users browsing this forum: No registered users and 34 guests