advertisement


Simple DC motor control

fox said:
Tidy layout. Just a suggestion: next version move the two voltage regulators centrally closer together and then the caps can be arranged sideways = lower height profile.
Click to expand...


The two regs need heatsinks and end up like this;



Underside. I've paralled the 10k resistors for the time being.


I use an old Axis as the workshop donkey.


As far as reworking the board goes, there's no chance of me doing as I don't have the skill necessary to do it. Ok, so the board will need a long enclosure if I put the tranny behind it but I can live with that. I'm quite happy with the board as it is.
 
Your images are not showing apart form the underside shot.

Its not the length, its the height, for other users. Those two PSU caps in a lower profile means it can more readily fit into tight spaces, like inside a turntable. Alternative is to add an extra bank of caps paralleled for lower profile. The VRs could also be rotated through 90 degrees and still use a heatsink.-- making better use of the real estate. V1 to V2 revisions are full of this kind of thing.

Like I said just a thought, for if the layout goes public for other people to use.
 
The motor I use runs at around 6.5Vdc for 33. Can the circuit easily be made to work for a motor that needs 6.5V for 33 and about 8V for 45? Would I need to do much more than up the voltage of the zener, tweak the pot values etc and reset the feedback values?
 
clivem2 said:
The motor I use runs at around 6.5Vdc for 33. Can the circuit easily be made to work for a motor that needs 6.5V for 33 and about 8V for 45? Would I need to do much more than up the voltage of the zener, tweak the pot values etc and reset the feedback values?
Click to expand...


I don't know Clive but I don't see any reason why the circuit can't be tweaked to accomodate your motor. Maybe Martin can expand on that.

As for getting some boards made up, I need to contact Paul ( PulseStudio ) as the layout is his property.
 
Yes, the easy way to get higher voltages is to set R1/R2 to provide a bit of gain at the opamp. Since the opamp will accept up to +/-18 rails it's be no problem to drive motors requiring up to say 15v to run.

Chopsaw's circuit is actually the one in post #1 ( the one in the spreadsheet does not work becasue the gain is inverting and with a +ve reference the output will try to go negative, ie sit at 0v all day long...)

I'll tidy up and re-post an updated schematic with some values and gain options soon.
 
martin clark said:
Yes, the easy way to get higher voltages is to set R1/R2 to provide a bit of gain at the opamp. Since the opamp will accept up to +/-18 rails it's be no problem to drive motors requiring up to say 15v to run.

I'll tidy up and re-post an updated schematic with some values and gain options soon.
Click to expand...

Can you also recommend a few options for the transistor when you update the schematic? You're being really helpful, thanks!

A concern I have about using gain in the opamp is that any noise from the voltage reg and zener will be amplified too. Whether this is a real issue is hard to judge.
 
We did use the other circuit prior to the one shown in post 1 Martin. I remember Paul building it and couldn't get it to work, because of what you said, ie the gain is inverting. He rejigged it and that first controller is still working well in my system. It's been fantastic really, never missed a beat.

I've emailed Paul and I'm waiting for a reply regarding the boards.
 
A concern I have about using gain in the opamp is that any noise from the voltage reg and zener will be amplified too. Whether this is a real issue is hard to judge.
Click to expand...
Well, yes, but as a linear reg it's far more than 'good' enough still - noise of a few 10s of uV when the speed regulation is probably uncritical to maybe 100-1000x that (say +/- 10mV). Its a motor it's driving, not a phono stage ;)
 
Chopsaw said:
We did use the other circuit prior to the one shown in post 1 Martin. I remember Paul building it and couldn't get it to work, because of what you said, ie the gain is inverting. He rejigged it and that first controller is still working well in my system. It's been fantastic really, never missed a beat.
I've emailed Paul and I'm waiting for a reply regarding the boards.
Click to expand...

Hi all

I have been away for a while and have just recieved a mail from Si (chopsaw) and a PM from Martin regarding these PCB's, I had them made by a local company that we ( our company ) use for all our 4 layer and above PCB's. I have a quote for 10x and 100+ and they are very competative IMHO, but if these boards need modifying then there will be an additional tooling charge which culd be costly, the price for the 10 up is £8 + VAT + postage and the 100 up will be approx £2.50 + VAT + postage. My original veroboard prototype was made using the very early design that Martin C kindly supplied to me, this as chopsaw has suggested has been working with no hiccups for some considerable time with his OL motor.

The reason for the revised circuit from Martin as I seem to recall was to try to make the resolution on the speed adjustment finer, however when I build the first one it didn't work so chopsaw reverted back to the originakl design, and yes there was a minor error with the reference voltage needing to be reversed polarity.

So I guess the bottom line is this, does this PCB need to be modified? and if so will the cost of the tooling for a new PCB be worth the hassle, as Martin has suggested "it is a motor controller and NOT a phono head amp" yes I believe that the basic design could be improved with the voltage reference device swapped out for something quieter (LM329) among other minor improvements, but like most things in life are the deminishing returns worth the time and cost implications, personally I don't think so, but that is just my opinion.

As it stands there is a huge voltage range that should cater for the majority of small DC motors out there, and with the adjustment of the sense resistor as Martin has suggested then the current PCB will suit virtually all aplications, I am perfectly happy to order these boards but would need an aproximate GB before going ahead, or if the die hards want to refine the board then simply mail me the new circuit and I will get a quote for the new revision PCB's...........................simpulzzzzzz

Paul
 
I would purchase a board. If someone wants to work on a V2 board I could live with a higher cost.
 
martin clark said:
You're a sucker for punishment, Paul ;)
Nice to see you back here
Cheers
Martin
Click to expand...

Yep I guess I never learn Martin, but hey I have been playing with my other passion (guitars) for the past few months, have just brought 3 vintage Ibanez Artists that are simply gorgeous, now that most vintage LP's and Strats are film stars salaries and I suppose a change is as good a rest so for my sins I will get a little involved here again.

Hope you are keeping well Martin and still have that amazing set-up we listened too back in the summer, Happy New Year mate and to all at PFM :)

Paul
 
Thanks for the kind offer of arranging, Paul.

Count me in, with or without revisions (like clive, I'm happy to pay extra for a 'v2').

Chris.
 
Very nice to have you back on PFM Paul ;)
I am up for a board as well in either form no problem with extra cost para me

Alan
 
337alant said:
Very nice to have you back on PFM Paul ;)
I am up for a board as well in either form no problem with extra cost para me
Alan
Click to expand...

Thanks Alan, I will try to get an overview of what is required here before committing to a GB on either revision, it may be possible to take the existing PCB and possibly have either an original Rev 1 or Rev 2 option.

OK I am no design engineer, but I can read a circuit and take instructions, so if anyone wants to come forward with a revised circuit that will keep the basic design but improve specific areas then fine, until I have this detail I can not re-design the artwork for the Rev 2 PCB quote, so come on all you bodgers send me your circuit mods, Martin did suggest a few up-grades like the following:

• Use a better voltage reference
• Filter the reference better
• a bit of brute-force HF filtering for the opamps supply
• Roll-off the gain of the opamp to increase stability (C3A is an example, needs playing with. Could be just a small cap between opamps signal inputs instead, which might work better here)

Whether these would have any real audible effect…I don’t know.

and then there is his final comment !! so if someone wants to come up with any improvements for the existing circuit then please mail your circuit to:

pulsestudio (at) hotmail (dot) com

and I will run it by one of our design engineers at work.

cheers
Paul
 
Paul - having built-up your board:

• Use a better voltage reference
Click to expand...
That's be nice, but not essential IMO.
• Filter the reference better
Click to expand...
This was one easy change - add a small cap to ground after the switch (1-10uF), which would make it a 2-pole filter.
• a bit of brute-force HF filtering for the opamps supply
Click to expand...
Add-in say 10R between the positive rail and the opamp supply pin/decoupling cap - the drive transistor already has a bypass cap (C45 - it would be nice to have a little more space for this, more like C41 or C42)
• Roll-off the gain of the opamp to increase stability (C3A is an example, needs playing with. Could be just a small cap between opamps signal inputs instead, which might work better here)
Click to expand...
Scratch that. A better answer - do what Werner suggested in Si's Origin Live thread, which is to add-in RC filtering in the feedback line from Rsense (R57) to the opamps +input. Say 1K series / 10uF to ground. That'd kill commutation noise above 16Hz, has no stability issues, could be linked-out and ignored, or optonally experimaneted with later.

Otherwise I'd only suggest making bits of the layout a little larger - where your PCB has footprints for the two connectors (for power LED and the speed select switch) it woul be simpler/easier just to allow PCB pins instead. Then it can be easly hooked-up with just wires into the holes.
 
You must log in or register to reply here.


advertisement


Back
Top