motor speed control

[cosmo]

Hi all. I attached a schematic and service manual descriptions of a motor speed control circuit for a Leslie model 60s organ/piano speaker. The motor doesn't work so I am asking for help. The service manual says to check transistors Q29 or Q30, or transformer T4 if the motor doesn't work. I checked all those and they seem OK.

As far back in the circuit I can trace is to the input primary of T4. There is no AC voltage happening here, and so I believe it is something a bit earlier in the circuit. I don't really understand the circuit too well. I think the motor is an AC motor because it says 24V 60HZ. So evidently there should be some kind of AC current supplied to the motor, but there is no AC or DC voltage going to the motor. I don't have an oscilloscope so I can't see what's happening anywhere except with the fluke multimeter.

Should I take out the circuit board and check all the components? Can I check all those with the multimeter? I don't know what the function of the IC unit there is, but it looks like it generates some kind of square wave or something. This is a circuit from the mid 1970's I think.

Thanks for looking and thanks in advance for any help or suggestions.

 

[Resqueline]

Starting at the beginning, measure the DC & AC voltages on all the capacitors. See if there's any change when varying the potentiometer. Report back with numbers.

 

[cosmo]

...roger that....will do. Thanks.

 

[cosmo]

motor control caps...

OK, here's the deal on the caps....

C68

This cap had a DC voltage of about 30.8V at the low pot setting and 23.4 at the hi.
It seemed to go up fairly linearly between low and high.

The AC voltage remained at about 0.2V from hi to low, which probably could be considered AC noise or something. However, the AC seemed to increase as the pot was being changed, but would settle back down.


C69

The DC voltage on this cap went from about 0V to about 8.3V from lo to hi. However, at about halfway up between lo and hi, it went up to about 15V. It also seemed to become unstable in the mid region, sort of blinking or something, but seemingly in a sort of random way.

The AC voltage on this cap went from about 0V to about 4V. It went up slightly changing the pot to about a quarter or a third of the way from lo to hi, then became unstable and blinking erratically up to maybe halfway or so, then at about halfway went to about 4V, where it stayed until all the way up.


C70

The DC voltage on this cap remained at about 13V from hi to lo. However, I could get it to go slightly above or below 13V on low setting, depending on how I went from high to low with the pot.

The AC went from about 0V to about 0.2V from lo to hi. However, the behaviour echoed that of C69, in that it became unstable and blinking about a quarter of the way up, and then stabilized at about 0.2V when reaching halfway up, and staying there.


C4 AKA C71;

I think the label of C4 is a misprint, as this cap is labeled C71 in other parts of the manual, which sounds more logical.

The DC component of this cap remained stable from lo to hi at about 1.6V.

The AC componenet remained stable at about 0V.


BTW, the service manual for this Leslie 60s unit can be found at
http://www.vintagevibe.com/t-tech-manuals.aspx

Thanks so much for your help! Let me know what you think, or what to do next.

 

[Resqueline]

Good job.
C68 seems to be at voltages I'd expect.
Voltages at C69 also seems normal. The 4V AC indicates that Q24 is working. The midpoint instability is likely due to a worn-out speed pot. Try to spray & jog it.
C70 voltage seems to be a plausible power supply for IC1.
IC1 is a divide-by-two circuit. For every pulse it gets from Q24 the output goes up or down (0.74V/0.24V), staying there until the next pulse. The output should stay at a random state every time you turn the generator off (which is when you'll be able to make that DC measurement).
C4/C71 voltages seems very low so there could be something wrong. Is R67 hot? It should dissipate 1.8W if it's ok. If so measure the voltages between collector-emitter on Q25-28 and across D6 & D7.
Page 25 (paper #) has a table with DC measurements btw.

 

[cosmo]

OK, thanks. I understand the circuit a bit better now, up until the output of the IC1 anyway. However, I don't really understand how the Q23 control works yet. I am supposing the voltage from the pot on the base determines how much current passes through Q23 to charge the cap C69 in time intervals proportional to the current, but I'm guessing.

I will check the output pin 3 on IC1 for those two hi/lo voltages and try to get them to change by turning up the pot briefly from the off position, or turn it up very low to see if it changes back and forth slowly.

I will check R67 for temperature. Is the power through R67 related to the AC current through the transformer T4? If so, I doubt if there will be current or power through R67 because there were no AC or DC voltages at the input of T4 when I measured there. I'll also try to measure the AC and DC across R67 as well.

It also seems strange that the voltage across C4/C71 is so stable. I don't understand the circuit very well past IC1, but I'm guessing transistors Q27 and Q28 are power transistors used to increase the AC current gain to power the transformer T4. It seems that cap should be passing some AC, just as a guess, becuase I don't know why else it would be there. An open circuit would do the same thing.

What happens when caps fail? Do they short out or do they become open circuits that hold no charge or whatever? Do they fail catastrophically or by measure?

I'll post those measurements on IC1 and R67, and try to do those measurements of the voltages listed on page 25. If R67 is dissipating power, I'll measure the voltages across the CE pins of transistors Q25, 26, 27, 28, and D6 and D7

Cheers.

 

[cosmo]

OK. I measured the output of the IC1 labeled number 3, which measured 0.8VDC WRT ground. This was a constant value when changing the pot up and down. Since you said the upper value was supposed to be .74V, it must be that the thing is stuck on the upper value. There was a constant value of 13V on the lead labeled number 2, and 0V on number 4, and of course 0V on the ground lead number 1. There didn't seem to be any AC voltages on IC1, WRT ground anyway.

R67 was indeed very hot to the touch, not being able to touch it for more than a second or so. It had 30V DC across it, and no AC.

I didn't have time to measure the voltages listed on p. 25 yet.

What would be the most likely of components to fail? IC1 or something else?

 

[55pilot]

Do you have access to an oscilloscope? You really need to take a look at waveforms on pin 4 of IC1 and on C69.

Assuming C69 voltages are good, you are down to two things. Either R59 is much less than 47 ohm (a short on the board, stray piece of metal, mouse pee?) or IC1 is dead.

R59 is easy to check and replace. You may be SOL on IC1. I do not know what the part number is and if you can even buy that.

There is a smaller possibility that Q24 is the problem. Its internal resistance could have increased so much that the peak current is not high enough to develop enough voltage across R59 to trigger IC1. You need to put a scope on that node to figure that out. If that is the problem, replacing R59 with a larger value (56 ohms would be a start) will do the trick.

If it turns out that IC1 is dead and you can not get a replacement, the whole circuit to the left Q26 can be replaced with a 555 timer, because that is what it is doing. You will have to analyze the existing circuit to come up with a resistance to frequency relationship and then design a 555 based timer circuit that implements that.

Good luck.


---55p

 

[cosmo]

Thanks for responses.

Alas, I sold my oscilloscope a while back to pay for school. I wish I had it still.
All I have left is my fluke multimeter. I could look for one on craigslist I guess. I got hit with all the layoffs lately, which is why I'm messing around with this thing in the first place.

However, by the multimeter, I'd say that IC1 is stuck on the high voltage state. I'm calling pin 4 the output of IC1, which was stuck at the high state of 0.8V. I guess I could check to see if there is any AC current on pin 3, which would mean some kind of waveform might be on it.

Which would mean the same waveform would be on R59, which I will also measure for ohms and so forth like you said.

Recall that there was an AC waveform on C69, which would mean there is a signal on Q24. But I guess it might be good to look at the output of that a bit closer, which seems to go onto the R59 and pin 3 of IC1.

If IC1 is stuck in the high state, should the motor be on or off. I am unclear how IC1 controls those transistor Q26 and the rest of the circuit to the right. Is this a pulse wave modulation PWM control on the motor? I am sort of learning as I go here.

If it is the case that the motor should be running in the IC1 low state, is there a way to short out something on the input of IC1 to get it to change state to see if the motor will start? ....if indeed IC1 is still operational.

A 555 chip sounds like a great idea. Or maybe just a stock motor speed controller. If you can point me to where to find a cheap one off the shelf then let me know. I'd like to continue to psych out this circuit and fix it, but I'd also like to get the unit up and running. Or, if you know of a circuit diagram somewhere that would use the 555.

Cheers.

 

[Resqueline]

It's likely that IC1 is defective, or that the input pulses are too weak to trigger it. The pulses going into it may be too narrow anyway to measure with the multimeter.
Nevertheless you confirmed there's something going on on C69 meaning the first oscillator does run.
IC1 output needs to alternate to run the motor, the quicker it alternates the faster the motor runs. Only AC can get through a transformer.
Q25-28 forms a low-power full-bridge circuit driving the first transformer which then drives the output transistors (which are in a half-bridge configuration), driving the output transformer.
Cap's can fail in many ways, not all will show, and much depends on the type and the application. They can short out (easy to measure), or they can get an increased internal series resistance. A cap meter usually can't measure the internal series resistance. Just add a similar cap in parallell temporarily to test for this condition.

 

[55pilot]

I agree with Resqueline with almost everything posted above. I agree that IC1 is likely the culprit, but I am going to disagree in that I am not ready to give Q24 a clean bill of health based on what you have posted.

Yes, C69 has an AC voltage which means that there is oscillation. But IC1 can be triggered only if the capacitor discharge happens really fast and causes enough current to flow through R59 to generate enough voltage. You can not be sure that this is happening till you put a scope to it. Resqueline pointed to this in the statement that the pulses may be too weak to trigger IC1, which will happens if there is too much resistance, either internal to C69 or internal to Q24.

Here's what I would suggest as the next 2 steps before you have to take it to the next level:

1) Replace C69 with a similar component. As pointed above, caps do age and die.

2) Replace R59 with a 56 ohm resistor. This may compensate for Q24 having too much internal resistance, if that is what the problem is.

At this point, if you can, you should try and replace IC1. I suspect that you will have a very hard time finding a replacement. I will move forward with the assumption that a replacement for IC1 is not available.

If those do not work, you need to move to the next level and for that you need access to an oscilloscope.

With a scope, I would look at the waveform on C69. Is it a sawtooth with a very sharp drop, or it is more of a triangle with slow charge and steeper discharge, but much slower than a precipitous drop.

I would look at the signal on pin 4. How wide is it? It should be a very narrow pulse. What is the maximum voltage? What is the frequency? What is the frequency at each extreme pot position and at the 25% 50% and 75% points.

When you know the pot to frequency relationship, then you can easily design the 555 circuit whose output is the same as the output of IC1.

On a totally unrelated subject, what in the world does this thing do? What is the purpose of this motor in the speaker box? Where is the speed control pot located and how does the performer control it?

---55p

 

[cosmo]

Maybe indeed Q24 is not doing all right,...either that or IC1 is not doing well. I think that the problem is pretty much narrowed down to that region though.

I measured the voltage on R59. WRT DC voltage, it was 0mV to 4mV, increasing linearly as the pot was being turned up. WRT AC voltage, it went from 5mV to 12 mV, increasing linearly as the pot was being turned up.

To a novice like me, this doesn't sound like the order of magnitude voltages that should be on that resistor, but I could be wrong of course. I don't understand that part of the circuit well enough to venture a good guess.

However, the circuit is starting to make more sense to me. e.g., the circuit seems to be divided into two components.,...the left of IC1 and the right of IC1, which are those full bridge things you all were talking about. I've been trying to find the time to study those more.

I think the RHS from IC1 of the circuit is working OK, and the reason is thus,... I shorted out IC1 pin 3 to ground, which is the base to Q26, which made the motor move a bit. I think this made the state of IC1 and the base of Q26 go to the low state. Then I repeated that manoeuvre in quick succession such that the motor moved at a very slow speed. Therefore, it seems certain that IC1 is locked into the high state for some reason.

I wish I had a place to hook it up to an oscilloscope. I looked on craigslist for a scope, but there were none offered. Maybe one will come up one of these days. I wish I still had my old one. I am between jobs right now so I can't just go out and buy one.

So, if someone sees an anomaly in R59 from those measurements, then let me know. Or, if someone knows of a simple way to interface a 555 chip to run Q26 by way of the 555 and the variable pot at the control, then let me know. If that pot there could control the 555 chip, which could in turn control the base of Q26, then things might be hunky- dory.

As far as what this unit does, it is a stereo speaker system that creates a doppler vibrato effect between two separated speakers, vibrato being defined as a slight wavering of musical pitch above and below the perfect pitch. One speaker wavers slightly above and below the perfect pitch, while the other speaker is 180 degrees out of phase, in that the opposite speaker's pitch is increasing while the other one is decreasing, and vice versa. It is meant to emulate electronically what a regular Leslie speaker cabinet does mechanically, which is to make a wavering doppler effect by way of a spinning cone arrangement over the top of a loudspeaker.

In this unit's operation here, I don't understand how the motor makes the doppler shifts happen, but I would LOVE to know. I haven't had time to figure out how it occurs, but it seems to have to do with rotating capacitors or something like that.

 

[Resqueline]

You could make a crude peak-meter add-on for your multimeter from a diode (1N4148) and a small capacitor (10-100nF).
The time constant of C69 dumping into R59 is just 1.55 microseconds, with an amplitude of up to 14V and a repetition rate of up to 120Hz.
1.55u * 120 *14 = 3mV ballpark average DC on R59 at top speed, matching what you measured. But this only tells us there are pulses, not their amplitude.
If Q24 has an increased internal series resistance then this would widen and flatten the pulses, making for a similar average DC voltage.
If you measure a good peak amplitude on R59 then it means IC1 is bad, but I'm sure we'll be able to adapt a simple two-transistor circuit like this to replace it.

Oh btw., the Space Generator here is unable to replicate the doppler shift of a Leslie speaker, only the phase shift.

 

[55pilot]

If you want to buy a low cost oscilloscope, eBay is your best option. There are some older models listed for under $50. If you are patient, you can frequently pick up a real deal. But you need to be careful and know what you are looking for because there are some crooks as well as some really clueless sellers out there. Whether you are misled out of ignorance or malice, you are still out the $$

If you want to make a 555 based circuit to replace everything to the left of IC1, you will need to know what the output frequency of IC1 is for the range of settings of the pot. For that you need to either analyze the circuit, which will be hard to do without detailed specs for Q24, or put a scope on the output of the circuit (the input of IC1).

Making a replacement for IC1 using a pair of transistors, as Resqueline suggested, is also an option. Come to think of it, it may be a better option. But in order to design that circuit, you really need to know what the input waveform looks like and for that you need the detailed specs for Q24 or put a scope on the output of the circuit. If you go the analysis route, without a scope you will have to assume that the circuit is working perfectly and there is no guarantee of that.

You will see a recurrent theme in my analysis: You need a scope. Maybe I have gotten spoiled by having all the latest and greatest equipment at my disposal and am unable to function without it. Some say that when you have the latest and greatest hammers in your tool-chest, you look for a way to solve every problem by putting a nail in it

---55p

 

[cosmo]

Thanks you guys. I put in a couple of bids on some scopes on ebay. There were plenty on there so maybe one of my bids will go through. Let me know if there is any particular brand or type to look for.

Thanks for the observation that the 'space generator' only varies the phases and not the pitch. I was really wondering hard about how it could change the pitch in an analog thing like that, but changing the phase seems much easier to do, although I will have to look at it again to see how it happens. I'm mostly worried about getting the motor to go at this point.

The manual said that IC1 is flipped by a pulse that looks like it comes from the C output pin of Q24, caused by Q24 A pin reaching a threshold voltage that builds up at the top of Q24. I wonder if by shorting out Q24 from A to C momentarily if that would cause a pulse that might flip IC1, if it is good. That way maybe it could be determined if IC1 is operational but just not getting a good pulse input.

Is it true that the 555 chip would replace the Q24 oscillator as well as the IC1 flip flop.? That way it wouldn't matter which one was bad, since it looks like the RHS of the circuit past IC1 is ok, since by shorting the output of IC1 made the motor move. I imagine the output of IC1 would go from 0Hz to maybe 60 Hz or so, since it is a 60 Hz 24V designed motor.

BTW, the motor control pot is on the end of an external cable that includes a wire also for the instrument input signal.

 

[55pilot]

Be careful placing multiple bids on eBay. If you win more than one, you are stuck buying all of them. The key to buying on eBay is to know what you are buying. If you do not know the product, do not bid. You may get lucky, but it is more likely that you will get burned.

You are right in your analysis of how IC1 flips. Since you measured a changing voltage on C69, we know that discharge is happening. What we do not know if the discharge is causing enough voltage to be generated across R47 to flip IC1. If there is too much internal resistance in Q24 or C69, it may cause the pulse to be shorter and broader. The lower voltage may not reach IC1's trigger voltage.

Yes, 555 would replace everything to the left of IC1 and IC1, so it will not matter what is bad.

---55p

 

[cosmo]

If the picture of the circuit shows up,....

This looks like the standard circuit for the output of a square wave of a 555 chip, where the pot controls the frequency of the wave.

f = 1.4 / [ (R1 + 2R2) × C1 ]

f = frequency in hertz (Hz)
R1 = resistance in ohms (ohm)
R2 = resistance in ohms (ohm)
C1 = capacitance in farads (F)

It looks like if R2 is max, then the frequency is low, and if it is zero,
the frequency is max.

Would this replace the LHS and interface into the RHS full wave thingies?

Recall that when I shorted the output of IC1 to ground the motor would move
a bit, so it would seem that the RHS is operational.

Given that the motor is AC with a label on it of 60HZ 24V,
would such a replacement work?

Question: Is the speed of the motor proportional to the frequency output of the flip flop?
Would the motor be zero speed at zero frequency? Would the motor be max speed at
60HZ output of IC1? Would the frequency be exactly the revolutions per second of the motor?

 

[Resqueline]

Yes, I guess it should work (well enough). Only two possible minor issues; will it stop completely when the pot goes open, and will the linearity be just like original. You'll just need to add a resistor in series with the output before you connect it to the base of Q26.
Motor speed (& drive voltage) is directly connected to the drive frequency but the motor may have a number of pole pairs and so run at 1/2, 1/3rd, 1/4th and so on [rps]. It may very well have a max speed above 60Hz, but presumably not much, because that also requires driving it with an overvoltage which would ultimately cause an insulation breakdown.

 

[55pilot]

In addition to the what Resqueline posted, here are a few other issues:

You also have to worry about the duty cycle. The duty cycle is the percentage of time that the output is high. The duty cycle is described by the simplified equation:

Duty Cycle = R2 / (R1 + 2*R2)

If you let R2 go too close to 0, the duty cycle will be close to 0. The reason R2 will never actually be 0 is because in the full equation the chip's internal resistances have to be included, which effectively limit R2 to be about 0.2 or more. The end result is that when your pot is at 0, the output will remain 0 and will go high for a very very short period, or the circuit may just not work.

To avoid this, you need to add a small resistor, maybe 100 ohm, in series with the pot. R2 then becomes the pot plus 100, so it can never go down too close to 0. If it turns out that you need to limit the maximum frequency, you can calculate the minimum value that R2 is allowed to be. In this case, the 100 ohm resistor I described becomes can be changed to this minimum value.

You still have to worry about the duty cycle. The motor is designed to work with a 50% duty cycle. The reason for IC1 in the original circuit was to make sure that the output stays high for exactly 50% of the time and low for exactly 50% of the time. You will have to do the same thing here. You will need to add a flip-flop like the 74LS74. Feed the output of the 555 into the clock input of one of the FFs (the 74LS74 has 2 on the chip). Connect the inverted output to the input and tie the set and reset inputs high to hold them in their inactive state. The output of the flip flop has half the frequency as the 555 but has 50% duty cycle.

What happens in this circuit is that every time the output of the 555 goes high, the FF assumes the value on its D input, which you have connected to the inverted output. When this new value is "latched" in the FF, the non-inverted output takes on this value and the inverted output takes its opposite value. At the next rising edge, the opposite value is latched in and the cycle continues, with the output changing state every rising edge. It really does not matter how much time the output of the 555 stays high and how much time it stays low. The output of the FF always stays high 50% of the time and low 50% of the time. Connected the output to gate of the transistor Q26 through a 1K resistor.

Finally, you have one more problem. You can not run the 555 and the 74LS74 the way they are running IC1. You will have to add in a linear regulator to generate 5V from the input voltage.

---55p

 

[55pilot]

I know that I am the one who suggested that you use the 555, but now I am having second thoughts. I would suggest that if the oscillator is working, you do what Resqueline suggested and use a pair of transistors to replace IC1.

Here's my reasons. You can not just use the 555. You had to add a FF and a voltage regulator and the circuit is getting bigger than the original suggestion of just "put a 555"

A much bigger concern is how the pot is wired. In the original circuit, the pot is a static device. The remotely located pot sets a voltage that is kept constant by the cap C68. In the 555 circuit, the POT is in the main RC current path. If you mount the pot away from the 555 (required in this device) you are just asking for trouble in terms of noise and picking up interference.

But to go to the IC1 replacement path, you need to make sure that the oscillator circuit is working and know the waveform on R59.

Sorry for misleading you.

---55p