Help Refurbishing Small Servo Motor

[vol.2]

I have this small AC servo motor that got sticky and noisy and I had to replace it. The motor is the drive motor for a cassette deck and it was generating audible noise in the music so was very annoying.

I was able to find an NOS motor, but it's actually a little bit noisier (in the music) than the original motor was before it started to go bad, so I think it's a case of some older components inside the motor needing to be replaced. So I opened up the old motor to replace, and I found that some of the caps are indeed not exactly on spec anymore.

There are 5 ceramic capacitors inside the motor. There are 3 times "103" (so should be 0.01uf I believe) caps that are wired directly to the motor windings, 1 times 2.2uf cap on the outside of the housing, and 1 times "005" (.005uf I believe) on the outside of the housing. The only cap that is reading close to correct is the 2.2uf one, which seems fine. The other 4 caps are as follows:

103 #1= 15000pf (should be ~10000pf)
103 #2= 15000pf (should be ~10000pf)
103 #3= 18000pf (should be ~10000pf)

005     = 3300pf (should be ~5000pf)

all three of the drifted caps seem like they have kind of high ESR

My goal here is two-fold:

1) replace the caps

2) try to improve upon the original specs of this motor by choosing the best replacement parts I can

My hope is that I can use parts that help to reduce the overall noise present in the audio circuit by improving filtering and motor stability. The original parts are generic ceramic caps from the 1970s, I can now buy better MLCC caps that I can easily use in this circuit (it never sees more than 6VAC), or even possible PET capacitors that are small enough to fit into the tiny housing.

Can anyone offer their opinions as to what kind of caps would offer the best functionality as replacements? Also perhaps if it makes sense to increase the values at all. Typically I wouldn't do that, but I'm willing to consider anything if someone has a good idea

Thanks

[Whales]

I don't know how your AC servo motors control their speed, so I'm probably not much use, but I might still be able to help.

Your capacitance measurements are not unusual even for new caps.  Ceramic capacitor capacitance depends on applied voltage, temperature and frequency.  Tolerances on these parts tend to be quite big, their primary market is for power supply rail decoupling where exact values do not matter.  I'll need to know their exact purpose and place in the whole circuit to determine if changing them will help or not.

Q1: Do you know exactly how your AC servo motors work and can you provide a circuit diagram of them (including the capacitors)?
Q2: Can you upload a recording of the effect this problem has on your music?
Q3: Is the distortion audible during silent sections of music?  What about if you press play with no tape inserted?

Wobbly pitch could be explained by mechanical problems such as this motor not maintaining speed, rubber bands losing elasticity or rollers being out of shape.  Noises that exist even when no tape is inserted (or when the music on the tape is supposed to be silent) would probably be caused by things other than the motor.  An actual recording would help here.

EDIT: I would also suggest keeping your mind open to causes of the problem other than the motor.  Example: the motor being unable to maintain a consistent speed could be a symptom, not a cause; the mechanisms it drives might be high in friction due to dried oils or too much oil.

[vol.2]

Thanks so much for your response.

Your capacitance measurements are not unusual even for new caps.  Ceramic capacitor capacitance depends on applied voltage, temperature and frequency.

I tested some other ceramic caps I had laying around that were close in stated values (not exact, but pretty close), and they read their correct values at least at one or two frequency ranges on my LCR meter. It does 100Hz, 120Hz, 1k, 10k, 100k. The values I gave for the caps inside the motor were the closest values that I could get to show up on the LCR meter at any it's ranges. It was enough of a discrepancy to give me pause

Q1: Do you know exactly how your AC servo motors work and can you provide a circuit diagram of them (including the capacitors)?

I don't know exactly, but I have worked on some larger electric motors and I understand the field and windings generally. Here's the schematic of the Servo and it's attachment to the motor:

Q2: Can you upload a recording of the effect this problem has on your music?

I could if you really think it would help after this, but I just did a quick spectrum measurement of it, and it appears to be a broadband hum generally between 50-200Hz, with a distinct 200Hz component. I was expecting to see 60 or 120, but neither of those seems to stand out. If you want an actual recording, how do you propose I share it here? I don't do a lot of that.

Q3: Is the distortion audible during silent sections of music?  What about if you press play with no tape inserted?

Yes to both. The noise is identical in both cases. If you turn the volume up loud enough to hear the noise, it is louder than any tape hiss noise, regardless of the Dolby setting (on or off).

Wobbly pitch

Pitch is rock solid looking at the output of a calibration tape on an oscilloscope.

rubber bands losing elasticity or rollers being out of shape

It has newish belts, a new pinch roller, and new idler tires. All the rubber is less than 2 years old, and I measured everything very carefully. I'm an old-hand at repairing tape decks.

the mechanisms it drives might be high in friction due to dried oils or too much oil

I don't think so. I fully cleaned it and lubed the right spots with the correct kind and amount of oil. I've done mechanical work on many tapes decks, and I have enough experience to not over oil or put it on the wrong thing.

[Whales]

Wow thankyou   Especially for the schematic.

I could if you really think it would help after this, but I just did a quick spectrum measurement of it, and it appears to be a broadband hum generally between 50-200Hz, with a distinct 200Hz component. I was expecting to see 60 or 120, but neither of those seems to stand out.

That's odd if you're a 60Hz country.  Perhaps your spectrum analyser has very poor bin widths at low frequencies?

If you want an actual recording, how do you propose I share it here? I don't do a lot of that.

You might be able to record the audio by plugging a laptop's mic in jack to the output of the cassette deck.  If this is too hard then even just recording it on your phone is good enough.  The forum should let you attach files (like .wav or .mp3) to your posts.  I can then have a listen and confirm your spectrum analysis.



Based off your description of symptoms I would not assume the motor to be at fault.  A bad motor wouldn't explain why the noise comes out of the audio outputs even when there is no tape being played.  Instead I'd assume something else in the circuitry to have gone bad (eg dried up capacitor in the power supply, pre-amps or similar).

Q1: What made you suspect it was the motor in the first place?    Is there something else, like the motor itself directly emitting a matching noise?

Q2: Are you listening directly to the output of the tape player, or do you have an external amplifier plugged in?  Have you confirmed that this external amplifier isn't the culprit?

If you have never diagnosed an audio circuit before then I can give a few pointers.  Start by measuring the main power supplies to see if they have lots of voltage ripple, if they are bad then everything else in the circuit will copy them.   Your schematic shows a discrete DC-DC converter, but not all of the device is using that, so I'd measure the voltage on the capacitors both before it (top-right of schem, C801, C802, C803, etc) and after it (top-left C804-C808). 

If you don't own an oscilloscope then in your case it's possible to cheat: grab a speaker, put a capacitor in series with it (100uF or so) and probe the power rails with it.  The capacitor will block DC but allow AC (such as the 50Hz/400Hz etc) through to the speaker.  If you can strongly hear the noise on the power supply then you've found a likely source of the problem. 

You will probably want to use a 100uF electrolytic or bigger so that not too much of the low frequencies are lost.  Make sure it faces the right way when probing the power rails (+ of cap towards + of power rail) otherwise it will fail to work as a capacitor and DC will damage your speaker.

[vol.2]

Thanks for the reply 

dried up capacitor in the power supply

I have already gone over every single cap in the whole unit. I ended up having to replace roughly 3/5th of them, but the rest look fine and tested very good. I suppose there are some freak occurrences of caps the look perfect and test perfect but are bad, but I kind of doubt it; I'm not a fan of shotgun replacing caps.

Q1: What made you suspect it was the motor in the first place?    Is there something else, like the motor itself directly emitting a matching noise?

Ok. I think I miscommunicated a little bit here. If you can humor me for a moment I'll try to explain.

There are two motors; the first motor is the one I'm currently working on. I know the noise is coming from the motor because it sounded good until the original motor went bad. The oil dried up and the brush contacting surfaces got too dirty and it turned into a resistor and started creating a big old humming sound. I was able to confirm it was the motor by whacking it and the noise went away; pretty cut and dry.

So I removed the motor and took it apart and cleaned it out and replenished the oil in the reservoir. I reinstalled the motor and it was more or less okay.

Then I found an NOS motor for sale and bought it and installed it. It sounded fine for awhile and then things started to sound noisy.

At this point, I'm only assuming that it's the motor getting noisy (I think it's a fairly good bet), and that the old motor refurbished (or new one, whatever) would sound better.

So fast-forward to now, and I'm going to put the old motor back together again, but I measured the caps and they seem out of spec based on what I have to measure them.

I can just put it back together, but I'd rather refresh the caps inside the motor, and maybe improve them a little bit. I reckon that some good MLCC caps might be a little lower ESR and more stable at the low voltages I'm dealing with than the original disc caps. Why not, right?

Now, you might be right, there might be a whole nother issue I don't know about yet. It's possible that one of the remaining 2/5th of the caps bit the dust unexpectedly in the past year or so, or one of the transistors went noisy.

Q2: Are you listening directly to the output of the tape player, or do you have an external amplifier plugged in?  Have you confirmed that this external amplifier isn't the culprit?

The amp is fine. I rebuilt it completely and it sounds great. I double checked the inputs with other sources and everything is clean and perfect. The noise is specific to the cassette deck, and it's very obvious that it's tape deck noise; for starters, it doesn't happen until I press play on the tape deck; just putting the amp on and turning the volume all the way up there is zero perceivable noise, it's a black hole

Your schematic shows a discrete DC-DC converter, but not all of the device is using that, so I'd measure the voltage on the capacitors both before it (top-right of schem, C801, C802, C803, etc) and after it (top-left C804-C808)

I rebuilt the DC-DC converter because it had a bad transistor. The other ones were perfect on a curve tracer. Again, all the bad caps were completely replaced.

If you don't own an oscilloscope then in your case it's possible to cheat

Sure of course I have a scope, but I also replaced all the bad caps already.

[andy3055]

That is not an AC motor.  It simply is a permanent magnet DC motor.
You need to clean the commutator and the gaps between the commutator segments. They may have dirt that is contributing to this. The speed control is done by the so called "servo" board components.  There is a speed sensor of some sort that has the violet colored wires. It goes in the motor on the other end. That signal is used to control the output from the transistors that power the motor. It is possible that one of these transistors or another component on that board is failing. All this is provided that you have ruled out mechanical problems and the motor bearings are lubricated. If you have a scope, monitor the signal coming on the vilot colored wires and go from there.
The caps inside the motor are for arc suppression at the commutator. Unless they are leaking,  they will hardly have any effect on the speed. If you have speed variations,  check the board. If it is motor  commutator noise,  clean the commutator and the gaps. Check the condition of the brushes also.

[vol.2]

That is not an AC motor.  It simply is a permanent magnet DC motor.
You need to clean the commutator and the gaps between the commutator segments. They may have dirt that is contributing to this. The speed control is done by the so called "servo" board components.  There is a speed sensor of some sort that has the violet colored wires. It goes in the motor on the other end. That signal is used to control the output from the transistors that power the motor. It is possible that one of these transistors or another component on that board is failing. All this is provided that you have ruled out mechanical problems and the motor bearings are lubricated. If you have a scope, monitor the signal coming on the vilot colored wires and go from there.
The caps inside the motor are for arc suppression at the commutator. Unless they are leaking,  they will hardly have any effect on the speed. If you have speed variations,  check the board. If it is motor  commutator noise,  clean the commutator and the gaps. Check the condition of the brushes also.

Thanks so much. That's very helpful.

That is not an AC motor.  It simply is a permanent magnet DC motor.

Okay, so where it shows the "VAC" on the purple wires, it's connected to the speed sensor component and doesn't drive the motor. Makes sense. The inside of the motor near the purple wires has a series of notches in the bottom on the outside, and what looks like a disc magnet on the axle inside of the notches. I suppose that induces an AC current that acts as a feedback network for the speed control circuit?

You need to clean the commutator and the gaps between the commutator segments.

I did this already. The brushes are carbon and appear to be great condition with lots of meat left on them.

The caps inside the motor are for arc suppression at the commutator.

How about the two caps on the outside of the motor? They kind of look like EMI suppression stuff; one of them is inline with what looks like a wrapped ferrite bead or some kind of coil. You can see a picture of it in my first photo.

If you have speed variations,  check the board. If it is motor  commutator noise

Speed seems totally fine. It's just too noisy. Hopefully when I put the motor back together and put it in, the noise will be gone, but I'd like to replace any bad components in the motor if it might make even a little bit of difference.

[andy3055]

The output of the violet wires is a square wave and it's frequency is the controlling factor for the speed. If your problem is mainly noise,  it can be electrical and/or mechanical. Your thinking of the cap outside is correct. Is the motor mounted with rubber grommets? If so, they may have become hard over time. You may have to run the motor with an external supply and see how bad it is and try to pinpoint the problem.  That is, if you can hear it.

[vol.2]

Thanks for the reply 

Your thinking of the cap outside is correct.

The one round disc cap in series with the ferrite bead is reading waaay off and seems to be high ESR, even for a ceramic disc. Should I replace it with a .005 MLCC?

Even if this isn't majorly contributing to the noise, it seems like it at the very least couldn't hurt? Then again, if the EMI suppression isn't working, could that contribute to overall noise in the audio?

Is the motor mounted with rubber grommets?

Yes, but the grommets seem to be okay. It's certainly still feels squishy and springy.

You may have to run the motor with an external supply and see how bad it is and try to pinpoint the problem.

In order to rule out the power supply creating the noise?

[andy3055]

Yes to all. This sort of thing is hard to pinpoint.  If you run the motor independently and eliminate any possible mechanical or electrical problems as much as possible, then you can proceed to other areas.
Do you hear the "noise" just or is it getting into the sound and getting amplified? If it is the latter,  look of poor ground connections that might interfere with electrical "screaming"

[vol.2]

Thanks

Do you hear the "noise" just or is it getting into the sound and getting amplified? If it is the latter,  look of poor ground connections that might interfere with electrical "screaming"

No, I can't hear anything coming from the device. All the noise is just in the audio. I did have an issue in the past with poor grounding on the foil sheets that cover the bottom of the PCBs, but I fixed that and they are okay now. Before I close it up, I make sure the foil measures continuity with ground

I don't know where else there might be a grounding issue to be honest. There isn't really and other mechanically connected ground leads, and the solder all looks good. I've been over all the solder joints in the past pretty carefully

I guess I'll just have to look around at stuff and see if I can isolate any sources. For now, I was just trying to ascertian whether or not it makes sense to replace the disc cap on the motor, and I guess I should, so thank you for that. I won't touch the three caps inside the windings that you say won't make any difference.

[andy3055]

It is hard to imagine what the noise is like.  If you can run a blank cassette and record it, probably you can post on YouTube.  Then put the link here. That way others can get an idea.

[vol.2]

Okay, thanks.

I will once I reinstall the motor. I'm waiting on a cap to replace the EMI suppression cap on the motor before I put it back together. It's kind of tricky to assemble/disassemble the motor housing, and I don't want to damage it by doing that too many times.

[vol.2]

Okay, so I tried this and it didn't work at all. I think ultimately the old motor (the one I was trying to refurbish) must have had some damage to the windings or something I was unable to detect, because when I finished working on it and put it back into the deck, it stopped working again.

I went back to the replacement motor, but before I reinstalled it I went ahead and took it apart and added some silicone based oil to the bearing reserves inside the housing and made sure the contact surfaces were good and shiny. I'm glad I did because the oil washers (it has small absorbent pads in the ends) were completely dry. After reinstalling the second motor, it now sounds much much quieter. It's still not a black hole, but it now sounds basically "normal"

[andy3055]

Everything you have mentioned so far points to mechanical noise as I can see. May be the old motor has worn out bearings in addition to some electrical fault. If it has markings or numbers printed on the outside,  take pictures and post here.

I am away for a couple of weeks.  So I may not be able to reply but others might. Good luck.