UK Vintage Radio Repair and Restoration Discussion Forum

UK Vintage Radio Repair and Restoration Discussion Forum (https://www.vintage-radio.net/forum/index.php)
-   Vintage Test Gear and Workshop Equipment (https://www.vintage-radio.net/forum/forumdisplay.php?f=36)
-   -   Mini-drill; Speed Controller Problem (https://www.vintage-radio.net/forum/showthread.php?t=141709)

Bazz4CQJ 27th Nov 2017 3:31 pm

Mini-drill; Speed Controller Problem
 
1 Attachment(s)
Coincidental to the thread which has been running on mini-drills, I’ve been looking at alternative circuits to the usual pulse-width speed controllers. The problem with those is that they are unable to compensate for loading of the drill. I found a number of circuits where the current drawn by the drill is sensed so as to provide a signal which is fed to a voltage regulator, so that when loading and increased current occurs, the regulator will supply increase the voltage, attempting to maintain the speed.

The circuit which is attached originates from this website http://quan-diy.com/projects/drillco...lspeedcont.htm, but I have re-drafted it to clarify what is going on. Unfortunately, my first attempt to construct this is not (yet) working; the drill just runs at full speed; maybe a problem on the PCB.

If there’s a problem with the circuit, it isn’t apparent to me and I’d appreciate any thoughts about its viability.

Thanks

B

orbanp1 27th Nov 2017 4:10 pm

Re: Mini-drill; Speed Controller Problem
 
Just a quick note.
The schematics does not show how the op-amp is powered (neither the web-page).
The LM1458 is not a rail-to-rail op-amp, the inputs should be higher by 2V than VEE, the minus supply to the op-amp, so you need to provide negative supply as well, below the ground potential of the power to the drill (the +12V - +16V).
If you want to power the op-amp from the same source as the drill (just the +12V - +16V) you need a rail-to-rail op-amp, like an LM358.

PS
I have not looked at the other aspects of the circuit, but there is a Spice model of the LM1084 on the TI web-site.

Good luck, Peter

Bazz4CQJ 27th Nov 2017 4:56 pm

Re: Mini-drill; Speed Controller Problem
 
Quote:

Originally Posted by orbanp1 (Post 994825)
The schematics does not show how the op-amp is powered (neither the web-page).
The LM1458 is not a rail-to-rail op-amp, the inputs should be higher by 2V than VEE, the minus supply to the op-amp, so you need to provide negative supply as well, below the ground potential of the power to the drill (the +12V - +16V).
If you want to power the op-amp from the same source as the drill (just the +12V - +16V) you need a rail-to-rail op-amp, like an LM358.

Thanks for that; it is not shown in the circuit diagrams, but actually the PCB feeds the input voltage to pins 4 and 8, so I think he has got that correct.

B

orbanp1 27th Nov 2017 5:16 pm

Re: Mini-drill; Speed Controller Problem
 
You missed the point in my reply that the LM1458 is not a rail-to-rail op-amp!
Powered like that, it will not amplify the current information from the parallel of R1 and R2.
The input to IC1A will be outside of its operating range! Do lookup the datasheet!
You need an op-amp like the LM358.

Linear has its own version of the LM1084, called as the LT1084, and LTspice has the model built in, no need to use the third party model from TI.

The other "anomaly" that comes to attention, is the value of the R5 resistor. In applications suggested by the manufacturers that resistor is in the 200 Ohm range, here it is 1 kOhm, which looks high!

EDIT: That value, given the load of the motor, seems OK. It would not be right without any external load!

Regards, Peter

Bazz4CQJ 27th Nov 2017 5:55 pm

Re: Mini-drill; Speed Controller Problem
 
Thanks for the comments Peter; I'm afraid you are dealing with a "old geezer amateur", who only recently commented that I really should stick to valve circuits! I've ordered in some 358's; I think the pinouts are the same, so could this work as a simple substitution?

Curiously, as we've discussed on the forum over the last year (starting with RF voltmeters) this looks like just another example of a circuit appearing in the public domain which is fundamentally flawed.

B

orbanp1 27th Nov 2017 7:15 pm

Re: Mini-drill; Speed Controller Problem
 
Hi B,

There is quite a bit of information is missing on that web-page.
That DC motor can have different versions (even when assuming that it is a brush-type motor), all having different characteristic curves (speed, voltage, torque, current). You could (and should) tailor the regulator characteristics to a particular motor.
The basic circuit as presented on that web-page indeed increases the output voltage as the load current increases, according to the simulations.
There is quite a bit of latitude in tweaking the component values though, and that is where simulation can help.

Regards, Peter

dseymo1 27th Nov 2017 8:15 pm

Re: Mini-drill; Speed Controller Problem
 
That was my first thought - perhaps this particular motor draws sufficient current off-load to drive the control circuit into 'flat out' mode. Maybe a few measurements and calculations would shed some light.

TrevorG3VLF 27th Nov 2017 9:09 pm

Re: Mini-drill; Speed Controller Problem
 
I currently use a 12V battery charger which is adequate to drive a DC motor running a coil winder.

I did an experiment with a rectified but not smoothed supply controlled by an SCR. An op-amp compared the motor voltage with an adjustable voltage. If the motor back EMF voltage was too low, the the SCR was triggered. Thus the speed was measured 100 times a second and corrected as necessary. I found that this was not necessary when winding thin wire so has been put in the round tuit pile. A battery charger and Variac is used at present.

Bazz4CQJ 27th Nov 2017 10:35 pm

Re: Mini-drill; Speed Controller Problem
 
I don't have a working knowledge of the simulation programs. This is quite a small drill, and off load, it takes 500mA @12V and 600mA @16V. Those figures are true irrespective of whether the op-amp chip is in the socket or not or whether the controller is in place or the drill is fed directly from the PSU. Maybe I should check the diode that sits across the regulator?

B

orbanp1 28th Nov 2017 4:04 pm

Re: Mini-drill; Speed Controller Problem
 
Hi Bazz,

If you do not want to wait for the LM358 IC to arrive, just give an additional -5V supply to the negative Vee pin-4 of the IC (instead of tying it to the ground), then it should work.

When the regulator chip works correctly, then there is 1.25V difference between the Output pin of the IC and the Adjust pin. Check it with a DMM.
You need load on the output for this to work correctly, because of the high R5 resistor (1 kOhm versus the suggested 200 Ohm - 240 Ohm).

Simulation shows that with the LED diode in the circuit the output voltage does not increase with load, if you short it out (no LED) then the output voltage increases with load.

And yes, the LM358 and LM1458 are pin-compatible.

Good idea to check out that diode across the regulator chip!

Regards, Peter

Bazz4CQJ 28th Nov 2017 4:49 pm

Re: Mini-drill; Speed Controller Problem
 
Thanks for that info Peter. I often use the drill with a small SLA battery, so that it is portable, so I'll use the 358 which should be here very soon.
I did wonder about the LED; suspected it was gimmicky. Certainly though, the idea of complementing a simple drill with a controller that compensates for loading should be very useful; just need to make it work!

B

Bazz4CQJ 5th Dec 2017 2:53 am

Re: Mini-drill; Speed Controller Problem
 
Well, it's working! Made the changes suggested by Peter (many thanks :thumbsup:) and I'm now "fine tuning" it. I also ditched the original PCB, which I though was a bit closely spaced in places, and finally got it working on a piece of Veroboard.

Experimenting by just rubbing a thumb against the chuck to load the drill and putting 15 volts in to the controller, the setting of the 50k pre-set is fairly tight; the "boost" is better turning the trimmer one way, but it tends to start "chugging" (pulsing?), but turning it the other way to get perfectly stable running seems to diminish the boost. Wonder if a little capacitance might be useful somewhere in there?

I'd like to do some comparisons, with the drill doing real work, between this controller and my pulse-width controller before I give it marks out of ten. I also need to try it out using the small SLA that I use when going "walk-about", e.g. fettling the cars, when the supply will be down to 12V.

B

Argus25 5th Dec 2017 4:42 am

Re: Mini-drill; Speed Controller Problem
 
Quote:

Originally Posted by Bazz4CQJ (Post 997026)
Experimenting by just rubbing a thumb against the chuck to load the drill and putting 15 volts in to the controller, the setting of the 50k pre-set is fairly tight; the "boost" is better turning the trimmer one way, but it tends to start "chugging" (pulsing?), but turning it the other way to get perfectly stable running seems to diminish the boost. Wonder if a little capacitance might be useful somewhere in there?

The chugging effect won't be cured by adding additional capacitance, though you might slow down its frequency. It comes about because as the motor is loaded, the sensed motor current drives the motor harder (speeds it up) so its simply oscillation from the positive feedback nature of the circuit. When there is just the right amount of feedback any friction/drag on the motor is just compensated for by increase in motor drive. The "chugging rate" represents the cumulative delays in the whole circuit.

There is a simpler version of this circuit, with a single OP amp driving a darlington output stage used as an emitter follower, the motor is in the emitter (or other transistor output circuit) and in series with the motor's wire to ground is a copper coil used as a current sensing resistor. The signal from that is filtered with an RC network and a proportion of it fed back to the OP amp's positive input.

It was a common technique used in very early JVC VHS video machines in the servos. So if friction slowed the head drum for example the positive feedback would compensate for that, its a technique of "artificial torque". A copper wire resistor was used to better match the temperature coefficient of the wire in the motor.

The other way to help stabilize the speed of a motor, if it has an attached frequency generator is to make a frequency to voltage converter and put that into a heavy negative feedback loop to control the motor speed, that way when the motor is forced slow, the feedback diminishes and it attempts to compensate. But most small motors don't have that generator attached, so the positive feedback method from a current sensing resistor is often much easier.

Actually I just remembered, I think I have a circuit at home, it was designed to speed and phase control a DC motor in a film camera, to lock it to video vertical rate (to avoid a rolling bar when the film camera filmed a TV screen/video monitor). It used an LM3900 Norton OP amps and the current sensing resistor trick for the speed control/artificial torque. I could find it and post if you are interested?

Bazz4CQJ 5th Dec 2017 4:00 pm

Re: Mini-drill; Speed Controller Problem
 
Hi Hugo, yes I'd certainly be interested if you can find that circuit. When I searched Google, I found numerous circuits based on pulse-width control, but only 2 where there was a feedback element. I don't have the expertise to design anything.

Continuing to assess what I have got with this one, it's working well on the SLA battery. I reduced the speed of the drill to about the lowest speed at which I thought it would do useful work and then measured the voltage being supplied as I loaded it. With no load it was 6V and that rose to ~7.5V when loaded, so it definitely is helping to maintain torque. I'm sure that modelling it in SPICE would help match the circuit to my particular motor. But even in its present configuration, looks like it will be superior to my PW controller.

B

Argus25 6th Dec 2017 8:35 am

Re: Mini-drill; Speed Controller Problem
 
1 Attachment(s)
Hi,

I have attached a circuit of the basic arrangement, you would just string a 1k to 2k pot across the supply to create Vin. If you want the voltage control range less simply lower the value of R3.

The 10k resistor with the * near it would just be adjusted so the 1k pot was about in the center of its range. This could supply a 2A motor without problems, it could also run on higher voltages like 20V

The circuit has filtering (anti hunt network) to reduce that chugging effect. When the 1k pot is set right, any drag on the motor is compensated so it doesn't slow down or speed up. The 0.6R current sense resistor was originally a coil of copper wire, you could use a small wire wound resistor of a similar value. One problem, there is often a lot of noise voltage from a brush DC motor.

You can't replace this OP amp with a generic part, the circuit is for a Norton OP amp, these work on difference currents at their input (not voltages) so its best to stick to the LM3900 op amp.

The darlington transistor arrangement could be two transistors like a BC639 & BD139, or you could use a darlington like a TIP140 or similar and save on parts and a resistor.

You can see that as the motor current increases on load the voltage across the 0.6R current sensing resistor increases, this is fed to the + of the OP amp and increases the motor drive.

You would need to build it and tinker with it to suit you exact motor.

The power supply connections to the OP amp are not shown.

orbanp1 6th Dec 2017 2:05 pm

Re: Mini-drill; Speed Controller Problem
 
Quote:

Originally Posted by Bazz4CQJ (Post 997026)
...
Experimenting by just rubbing a thumb against the chuck to load the drill and putting 15 volts in to the controller, the setting of the 50k pre-set is fairly tight; the "boost" is better turning the trimmer one way, but it tends to start "chugging" (pulsing?), but turning it the other way to get perfectly stable running seems to diminish the boost. Wonder if a little capacitance might be useful somewhere in there?

...

Hi Bazz,

Glad to hear that the circuit is working fine!

Regarding placing a small cap somewhere there for stability.
There is no use for that in this circuit!
The circuit for the current monitoring is a positively fed back circuit with a phase shift of 0 degrees.
The only thing that prevents such a circuit from oscillating is that the loop gain is less than 1, or 0dB! That one you adjust with that pot.
Absolutely no need to tinker with phase shift with a small cap in this case.

Regards, Peter

Bazz4CQJ 7th Dec 2017 2:22 am

Re: Mini-drill; Speed Controller Problem
 
Thanks for that additional design, Hugo. I'll maybe take a look at that over the Xmas break.

B

Refugee 7th Dec 2017 2:49 am

Re: Mini-drill; Speed Controller Problem
 
I once thought up a concept for a speedometer for a model railway.
The idea was based on the fact that a DC motor with brushes does not draw a constant current but actually has a waveform that can be seen on a scope across the low value resistor in series with the motor.
If an amplified waveform could be compared with an oscillator or a timing circuit the speed could be controlled quite accurately regardless of loading on the motor.

Bazz4CQJ 7th Dec 2017 9:48 pm

Re: Mini-drill; Speed Controller Problem
 
Does the loading on a model engine change very much as it moves around the track? It's not an area where I'm up to date with what kind of power supplies and controllers are being used - I vaguely recall reading about various ways they been improved since the days when it was purely a variable transformer. I think that in the case of the mini-drills, which are not geared down and typically run at quite high speeds, any loading rapidly reduces the torque and they will easily stall. Hence my interest in going away from the usual PW controllers, which have no feedback system, to the design discussed here which will try to keep the drill turning even if the load increases appreciably.

B

mark_in_manc 7th Dec 2017 10:53 pm

Re: Mini-drill; Speed Controller Problem
 
I've been reading this thread with interest. I'm currently working on a big DC motor rescued from a scrap treadmill - max 20A+, 130v, 3-ish HP - for application to a machine tool. I've run it using a cheap-ish Chinese 230v PWM motor control module I happen to have, fed from a Variac, which claims it can deliver 8A (the module, and the Variac!). I find that when I load the motor, the volts across it stay the same, but the current rises steeply - and the speed is maintained. Maybe this controller has a feedback loop built-in. I have two others bought cheaply on-line to experiment with, which claim much higher current delivery - I'll post back on how I get on. I was wondering about paralleling-up the output transistors (scrs, triacs?) on device #1 rather like in my Farnell stabilised DC bench supplies, since the control side of it seems to work rather well. I guess I may have to trace out the circuit.


All times are GMT +1. The time now is 9:10 pm.

Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.