Confused by strange speaker movement, EVM-12s - Randall RMS-150 combo

[CarsonM]

This is my first time posting to this forum so please point out any posting missteps.

I recently installed a used EVM-12s into a Randall RMS-150 combo amp (It's a 120watt solid state amp with a vented cab) and though the speaker sounds great in the amp I noticed that there is quite a lot of extraneous cone movement. It looks like every time I play a note (even the highest note on the guitar) the speaker will move like I'm putting a 2hz sign wave into it at 300 watts and most of the movement happens as the note is dying out and not during the initial attack. I can't really hear anything strange happening when it does this. The movement is almost slow enough to count the cycles and appears to be full cycle with complete in and out movement. I understand a little about speaker efficiency and why the EVM would be capable of more extreme excursion but I don't understand what could be causing it in this instance. Another detail is that I have two of the exact same amps so I checked whether the other amp caused the same movement, and it does, but only with the one speaker. This doesn't happen with any of the other three speakers I played the amp through. Just the EVM. The other speakers were a 200w celestion bass 12", JBL G125, and a Fender 100w.
I tried searching around for different speaker symptoms but was unable to come across anything similar relating to speaker failures, but there might be a term that I don't know. I'm going to check the speaker with another amp tomorrow. I suppose if another amp causes the weird movement then I can assume that the problem is with the speaker. It's highly possible that Randall produced an amp that had some issue in the output section, but I know my Traynor YBA-1 wouldn't have the same problem.

I just got in a new oscilloscope so perhaps I'll check the output of the amps for sub oscillation just for giggles, though I doubt that's the problem. I don't know if any oscillation would be visibly apparent in the speaker, and only one type of speaker at that.

Thanks for trying to make sense of this

[Jon Snell]

I would guess that the frequency response of the amplifier is too low, verging on instability. There should be nothing getting to the speaker below 30Hz at full power, otherwise it is just a waste of energy and can damage the speaker.
Just like some amplifiers claim to work between 10Hz and 50kHz. We can't hear those extremes, so why waste efficiency.
If you post a schematic diagramme, we can tell by the components used.

[pdf64]

I suppose both amps may be of a similar age, so it’s possible that both amps have got tired ecaps, in their power supplies etc.

[loudthud]

Is it a RMS-150 or RSM-150 ? The only listings for a schematic of the RSM-150 I could find were $15 to $20 USD.

Anything like what was discussed in this thread ?: https://music-electronics-forum.com/...te-and-digital

[J M Fahey]

Just 2 possibilities:

1) amp is on the verge of oscillation.
If it were full, it would motorboat; if near but not fully, damped oscillation may be triggered by normal guitar signal, exactly what you describe.

Typical cause is way too tired supply caps,so poor decoupling.

Not only the main ones, but also smaller ones decoupling different preamp stages.

2) speakers have 2 "springs" keeping voice coil "where it should", not only side to side but front to back.

One is speaker edge, EVM should have a soft-ish impregnated cloth edge, but not a fluffy foam one like on Hi Fi woofers.

Internal "suspension/spider" is a corrugated (typically orange but may be tinted black) cotton cloth disk impregnated in phenolic resin and hot pressed.
Itīs joined to cone and voice coil using a bead of adhesive all around.
Sometimes this adhesive cracks and leaves cone and voice coil free to move a lot back and forth.

[TomCarlos]

Welcome Carson....

I am posting an image of the Power Amp section. As I mentioned in a prior post about the RMS-150, Music Parts dot com does not want you to reproduce the entire Service Manual. So I hope this image of the Power Amp section does not get me in trouble!

Tom

[TomCarlos]

Power Supply....

[J M Fahey]

Well, not much of a secret there, it is the *classic* 100W Randall Power Amp.
As shown, schematic has 3.3uF and 10k at the input,which is 5 Hz.
You might reduce that to 1uF which is still 5 Hz, still well below guitarīs 80Hz , might even use 0.47 uF for 30 Hz (I would), but in any case, the amp worked well straight from Factory, like Enzo often says , so search somewhere else.

Again: Main caps and preamp +V rail ones.
Suspect preamp uses fets?

Like an SS version of classic Tube circuits?

Those can motorboat as well as any Tube

[TomCarlos]

Preamp ...

[TomCarlos]

And for those who want the correct wiring of the Footswitch, visit the thread Randall Switchmaster RMS 150 - Both Channels Option - Schematic Request

[CarsonM]

Thanks for the responses everyone.
Yep, the problem is most definitely the amps.

I'm not sure what the rules are for posting the schematic here. I can't tell if it's being sold under some legal exclusivity or not. The seller is in the business of selling manuals of all kinds on ebay. I do have a copy though. Perhaps I'll snap small sections of it.

I have replaced the ecaps on the output board and in doing so I noticed that C32 on the schematic that Tom shared was installed in reverse on two of the three (two combos, one head) of these amps I have right now. I'm not savvy enough with solid state amps to spot if it was a schematic error or a manufacturing error but the schematic seemed correct to me so I installed those caps as listed. Positive side tied to output.
I also replaced the V+ caps on the preamp board but have yet to replace the mains can caps partly because I was dreading trying to figure out how to mount the caps since I can't find twist lock caps of the same values. The caps are 6000uf 50v for the 40v rails. I'm gonna order caps today and just be done with it. With tube amps I'll occasionally up the filter capacitance by a tad, are there different considerations for solid state amp power supplies? It can actually save me a bit of money to fudge the values a bit. 6800uf 63v Illinois caps are just $2. Or I was considering opening the cans and rebuilding them using a couple smaller caps so I don't have to devise a new mounting situation, though the amp is crowded in the way they are mounted currently.

[TomCarlos]

Carson.... Remember that back in the day, 6000uf might have been readily available. And those caps might have had a big swing in tolerance. So a 5600uf is "close enough." The others here can weigh in on that. If you can find radial caps, you can drill some holes in the lead traces and mount them that way. You can also use a dab of silicon to help the cap adhere to the PC board.

[CarsonM]

I'm thinking I'm gonna get rid of the can cap harness all together. Mount the rectifier to the chassis and add a terminal strip so I can use modern radial options. It'll add some distance between the caps and the preamp board and make the whole section more serviceable in the future.

[loudthud]

The low frequency instability that J M Fahey spoke of ( post 5) would be caused by C42. In some earlier Randall amps, the power supply for the preamp was zener regulated ( that would prevent the instability ). In this amp, it looks like they just copied the earlier schematic with everything running off of the same supply node, so instability is a real possibility.

Definitely investigate the situation with C32. Ohm the plus and minus leads to the output and the -40V supply. The schematic posted above I believe is correct, plus side to the output, minus side to the two resistors. This cap will cause all sorts of crazy problems if polarity is not correct.

[J M Fahey]

Good eye: feeding that many gain stages from a single node is crazy, even if using larger-than-normal C42 (1000uF instead of typical 100uF).
Q1-2-3-4-5-6-7-8-9-10-11?????? WTF?

It must have worked out of factory (or it wouldnīt have reached dealerīs shelves) just by the slightest margin, depending on high quality caps.

I would start by replacing C42 with good one (any modern "good family" one will do, say Panasonic, Nichicon, etc.), but also do a little brain surgery.

As Enzo says, usually re.engineering is not justified ... this is one of the few cases where it is.

Please post a couple Preamp PCB pictures, one from above showing components, specially Q1 and Q5 area (and a couple inches around) also showing R4 and R29, and same from below, to identify the track coming from "A" and feeding top of those resistors.

There must be a track joining both.

I would cut that track section (1-2 mm wide is enough,with a sharp utility knife and removing the slice of copper) away from the rest of the "A" line, scratch both sides to expose copper and bridge the gap with a 1k resistor and connect this separated section to ground using a 100uF or 220uF x 50V electrolytic.

That should solve it for good.

I noticed that C32 on the schematic that Tom shared was installed in reverse on two of the three (two combos, one head) of these amps I have right now. I'm not savvy enough with solid state amps to spot if it was a schematic error or a manufacturing error but the schematic seemed correct to me so I installed those caps as listed. Positive side tied to output.

Output is basically at 0V DC, what do you read at the other end of C32?
I would expect some -20V DC , please confirm.

Recheck measuring straight across cap legs/pads that "+" is more positive than "-"

An uninformed previous repair guy might have thought "hey, electrolytics have negative connected to ground" or something.

The caps are 6000uf 50v for the 40v rails.

Anything from 4700 to6800uF x 50/63V will work fine there.
Modern ones will be physically smaller than originals so easier to fit.

Drilling PCB (and scratching solder resist around new holes) as suggested is fine.

C28 is "backwards" but given the microscopic voltage there (40mV) it does not matter.

Hope speaker out jacks are plastic or isolated from chassis, their "ground" terminals returning to PSU ground through their own wire, to keep chassis metal "clean" as far as grounds go.

If possible also post a closeup of them.

For a "metal" amp, it does not have much gain at all, thatīs why they could work even with poor-ish decoupling.