Trying to fix loss of volume in a 1978 Randall RG300 Head (SS)

[ParthaD]

1 more pic..

[ParthaD]

Just one quick addition, you don't need to bring new caps up slowly.

Thanks for that note Jag... as I got deeper into this I did read that it is common practice to condition only NOS stuff older than 5 years.

[J M Fahey]

Hi ParthaD.
Sorry if I did sound rude, I was afraid that some avoidable damage could be done to your fine amp. It's *very* easy to slip a test point, touch something else, and turn a reasonable repair job into a nightmare. It's happened to me more times than I want to remember.
The schematics I found are real small but readable, they will do.
Excellent pictures, very useful.
Consider this amplifier as a very loud "Twin" substitute, obviously SS, sharp and snappy (not exactly sweet), needing some good external pedal for distortion, with classic reverb and tremolo effects.
It has no effects loop, but you can easily split it into pre and power sections.
Go to picture called "pots3".
On the upper side, at 1/3rd left to right, you'll find a pot that can identified by: it has a pull switch attached, is in front of a grey 470uFx35V electrolytic capacitor, and has, from left to right: 1 green wire<>1 beige/light tan<>1 tan+1 white, it's the master volume pot.
Desolder these last two (tan+white) and tape them for safety.
Get a guitar cord, leave the plug in one end, and on the other desolder the plug, solder the screen to the green wire on the master pot and the hot/live/center wire to the free pot lug.
Now you can send audio straight into the power amp, bypassing the (suspect) Randall preamp.
You can use the line out from another amp or some pedalboard. An MP3 player will be weak there.
I suspect your amplifier should work properly.
Post your test results and we'll move on from there.
Good luck.
Juan Manuel Fahey.

[ParthaD]

Amazingly helpful response Juan..!!

No offense taken Juan, I understand the spirit in which the previous cautionary advice was given!

Thanks for the schematics and your detailed advice... will post results back tomorrow... thanks again!

[ParthaD]

Pot lug question...

Hi ParthaD.
Go to picture called "pots3".
On the upper side, at 1/3rd left to right, you'll find a pot that can identified by: it has a pull switch attached, is in front of a grey 470uFx35V electrolytic capacitor, and has, from left to right: 1 green wire<>1 beige/light tan<>1 tan+1 white, it's the master volume pot.
Desolder these last two (tan+white) and tape them for safety.

That pot has 3 lugs at the top and 2 at the BOTTOM also (I guess for the pull boost). The white wire from the top right goes to the bottom left lug and there is a tan wire that comes out from the board to the bottom right. When you said to desolder the white and tan wires you meant the top ones right? Should I be leaving the tan wire going to the bottom right lug as it is? Thanks!

[J M Fahey]

Yes, the three pins on top belong to the pot, the two below to the extra-gain switch.
You have to leave free the top right pot pin, desoldering the white and tan wires . Keep them soldered between them and wrap that junction with a piece of tape, just because of Murphy's Law.
You will solder the "hot" (center) wire from your new audio input cable to the free upper right pin, and the shield/ground to the opposite pin, the one that has a green wire which is *not* desoldered, it carries the ground. Now you can inject some music into tour power amp, I guess it will work. If not, at least we are getting closer to the real culprit.

[ParthaD]

News is good!

Hi Juan... I did not do exactly quite what you had asked (shame on me but hear me out please.. ) but I think I established the same facts... I had some Radioshack wires with alligator clips lying around and a loop machine so basically instead of desoldering those lugs I simply connected the alligator wires as instructed (if that didn't work I was going to do the desoldering and soldering next but I figured that even with the signal attenuation that was bound to happen with the other wires to the preamp section still connected - the ones that you had asked me to desolder - I should still get a significant boost).

I adjusted the output from the loop box to match the AC mV output of my guitar (about 50 mv) and turned the amp on and presto! I had HUGE VOLUME, couldn't put it past 3, things started shaking, previously I had to put it to 10 and even then it wouldn't get this loud! I was sending it to a crate 4x12 cab with a 4ohm impedance! I can only imagine what this baby puts out with TWO such cabs and 300 Watts at 2 ohms!!!

I also noticed the o/p transistor got pretty hot (could still touch it) and after turning the amp off I also touched and found the 3 op amps in pic 96 getting warm too.

So this establishes that the power supply transformer, filter caps and the power amp are fine, right? Also even without doing these connections I did get some sound out at max volume connecting to either channel. Therefore, would I be correct in concluding that the preamp problem is bound to be AFTER the tone stacks of both channels come together otherwise the audio problem would only happen on 1 of the channels (since the 2 tone stacks appear to be separate over a large section before coming together both in the schematic as well as looking at the amp physically) and just BEFORE it enters the power amp section?

So if I insert the same signal using only the hot clip (the ground is common anywhere right) at successive points I could probably arrive at the source of the problem. But before I do that I thought I'd seek advice and since you know this circuit so well perhaps some suggestions on where I might insert the signal next physically in the amp since I am still not sure I can translate the schematic to the physical part yet.

Thanks for your help, the amp sounds simply awesome!!!

Btw for those interested I found a very good explanation for the old RG series circuit at this url:

Randall RG-80/100 series: Harmony Central User Reviews

[J M Fahey]

I'm happy the most expensive and dangerous parts, the power supply and power am boards work perfectly.
You have been empirically closing the net, congratulations. Your drum machine is a very good signal source.In all, you have managed very well.
Tomorrow I'll suggest some signal injection points. As you see, and as Enzo and Bill52 had already suggested, it's *much* faster to inject signal and either follow it (if there's absolutely no output) or proceed back, step by step, until finding the culprit stage.
*There* you start measuring voltages looking for abnormal values.
Good luck.
Juan Manuel Fahey.

[ParthaD]

Channel 1 hacked to fully functional now!!!

You have been empirically closing the net, congratulations.

Thanks Juan, I feel like Luke Skywalker thanking Yoda!!!

Got hooked to this signal tracing business... have some more to report. I decided to take a shot at tracing the signal on the channel 1 side. I connected the signal into the hi channel 1 input jack and did not short any leads from there. Instead I followed the pattern you showed me on the master volume pot and tried connecting the rightmost lug of the volume, treble, mid, and bass pots in turn to the same master pot lug and they all checked out, i.e. I had loud sound. So I went further (navigating slowly towards the master pot) and ultimately came to a certain resistor downstream of the volume/tone stack and when I connected that to the master pot I actually had a fully functional channel 1, meaning the volume, mid, bass, treble all actually worked normally so I had dual volume control, both from the channel volume and master volume and could go as loud as I wanted!!! Have attached a pic of that. When I connect to the other side of the resistor I lose the signal. However I am confused because the resistor did not read open, it read 73.5K ohms (measured with the amp off). However until now I've always found that connecting to either side of a resistor works if it works on one side and this is the first one where it was not true. Will try to google and find the actual value based on the color codes on this resistor.

During the course of this evening, also looked up the NPN transistors all over the circuit, the TIS98 ones - which is a Fairchild TO-92 xstr - found the datasheet - and read on the net how one can check it with a meter, reading resistance when connecting one lead to the base and checking emitter and collector in turn - they should read equal values - of course did this with the amp off. They all appeared to check out using this method.

Just fyi, if I connect the loop jack I've been using to test today, into channel 2, it won't work yet because it probably joins this side of the preamp somewhere AFTER the point where I am pulling signal from.

Getting there slowly!

[ParthaD]

When I connect to the other side of the resistor I lose the signal. However I am confused because the resistor did not read open, it read 73.5K ohms (measured with the amp off). However until now I've always found that connecting to either side of a resistor works if it works on one side and this is the first one where it was not true.

From the schematic this appears to be R12, 68k which is close enough to what I read, and it goes to ground, hence I lose signal here! Right?

[Enzo]

You cannot make a rule that signal will always go in one end of a resistor and come out the other. In electronics, it is all about the circuit, not the parts. A resistor can do many things all depending upon how it is used.

Consider a piece of wood. If it is horizontal, it is a shelf, but if it is vertical, it is a support leg. In your circuit there is a signal path through the amp. In many resistors, and other parts, the signal does indeed go in one end and out the other. But in other cases, such as R12, the resistor is in a support role. R12 goes to ground. There will not be any signal on ground. The other end is electrically close to Q3, and will have signal sitting on it.


Try to envision a signal path through the amp. For me, I like to hold a schematic at qarms length and squint my eyes to dim out small details. I find I can get a bettter idea of hte overall flow of the circuit.

Without getting too detailed, your signal path is something like this:
In through C1, then through Q1, Q2, C3, the tone controls (tone "stack"), through a couple resistors to C8. Note that channel 2 is really the same circuit, just a few details are changed or added to make it sound different. Note also channel 2 has the couple resistors going to C8. C8 is where the two channels join and the rest of the signal path is commmon to all.

Those couple resistors we just mentioned would be called mixing resistors, after their function to mix the two channels together.

C8 feeds the gate of Q3, which seems to be where you are. The signal leaves Q3 and through Q4, then C12, and the master volume control. From there it enters the power amp through C15.



So if you have signal at R12, one would hope you also had it at the bottom end of R13. That point is also directly connected to the base of Q4, so look at the output end of Q4. Is there signal at the non-ground end of R15? Then at the other end of C12?



By the way, I would be highly suspicious of those 3.3uf small electrolytic caps. C2,3,12,15,25,27. And any others like them. WOuldn;t surprise me to find C12 bad.



In your photo, it LOOKs like you are on one end of a 100k resistor, and it is connected at the other end to a 10k. That sounds like the mixing resistors. But the image is hard for me to see, and I could be way off base.

[ParthaD]

That was not R12...

... Enzo, I made a mistake, in that last pic that I posted that resistor was not R12, it was in fact R9 which is 100k (brown/black/yellow) and the one on its left is R92, the 10K (brown/black/orange) . And last night I was testing very late at a very very low master volume so when it appeared that the signal had gone away across it it really hadn't, the signal just drops significantly after passing the 100k resistor but that is normal based on the test point signal voltages shown on the schematic in rectangles, from 55mV to 20mV.

Have also found the equivalent jumpering point in channel 2 i.e. when I take the signal from R47 (100k) and connect to master pot, I get full volume on channel 2 also including the tremolo effect. So the problem appears to be in the section from C8 to the Master pot.

Will keep digging. Thanks for all your help!

[Mars Amp Repair]

Well,
It looks like you have all kinds of good advice for this amp...thought I'd just throw in another.

All those gray looking electrolytic capacitors should be suspect after this many years. they can be used for filtering (usually the larger ones) or for coupling the signal from one stage to another...the latter, your audio signal tracing will ferret out.

the other less obvious function is for something called 'emitter bypassing'. I won't go into the reasons, but just know if a cap used in this function is either shorted or open, you will have diminished gain in that stage.

It might not be so obvious with the audio signal tracer as we don't always know what type of gain each stage 'should' have so we have no point of comparision.

You can take a known good cap (minding the polarity) & just jumper it across the caps in the circuit one by one while listening for the effect.

This will prove out the old cap only if it is either lost some of it's capacitance or it's open.

If the cap is shorted, this won't help you, however if you know what funciton the cap is used for ie:coupling, bypassing, filtering, a voltmeter can help you then.

If it is used as a emitter bypass funcion, you can read the + side to ground & expect to get about 1 - 3V on the + side & the other side typically goes to ground.

If you get 0V then it's possible the cap is shorted. In this scenario, you will have greatly diminished output...possibly distorted, too.

If it is a coupling cap, then you should have a different voltage on one side than the other. Typically higher on the preceding stage...possibly 10-20V on the + side & ususally much less on the other side. Not a definite...just a rule of thumb.

NOTE: do not try this method in the power stage. They are typically very intolerant to cap jumping, which can cause power surges & possibly damage the power amp stage. there are typically only a few small electrolytic capacitors in the power stage of older solid stage amps like this one, so best policy if you suspect them is to just replace them, as you can spend hours trying to prove them out.

Anyway, that's a lot to take in at one time, anyway. glen

[ParthaD]

Well,
It looks like you have all kinds of good advice for this amp...thought I'd just throw in another.....
....
Anyway, that's a lot to take in at one time, anyway. glen

Thanks for the detailed reply Glen! I will try to use all of that in my ongoing troubleshooting. Have been busy with other things so haven't reported for a while... what I did find on the web was a detailed description of the circuit by someone who had worked on this extensively in the past. This person (Dave) had done a review of the amp in 2009 that I had reported before but I found an earlier one by him from 2005 which went into the circuit itself. I have listed both below:

2009 review

2005 review (more detailed)

Based on this 2005 review it could be that one or both of the protection diodes D3, D4 might have blown during a past abuse! It is right between the point where I have traced signal, and the master pot, i.e. after the 0.1uF C8. These are the 2 orange ones right above the thick ground bus in the center of picture 105 in line with the 4th pot from the left. Will try to verify over the holidays and report back.

Thanks for all your help!

[ParthaD]

May have found the problem!!!

Am posting a couple of pics to help you understand what I found. First to give you the immediate result....

The original schematic that was posted by Juan was split in 2 pics, I spliced together the relevant sections, left some overlap in the image since the left and right side were at slightly different magnifications. You will see that the diodes D4 and D3 (left side) are shown in series in terms of direction. However, on the board they were actually in parallel (the black lines on both were towards ground) and appeared to be a kind of overload protection. This fits in exactly with the info provided in that amp review by Dave that I had mentioned in my previous post. I have been kind of suspicious of these 2 and been checking them for continuity in both directions all along (by reversing my meter leads and checking with the amp off of course, one of them showed the good results, i.e. open only in 1 direction but the other was flaky, it would sometimes show open both ways and sometimes not).

I am also posting a pic showing the suspected components.

Today as I was testing suddenly the pretty loud sounds that I had got before abruptly got cut off! So I immediately checked these diodes and found that D3 was now ALWAYS open both ways! D4 still checks out ok. There were 2 other identical diodes fairly close in another section of the circuit and here too I found 1 of them open. It was D5. However, in this case (based on the earlier mentioned Randall review info), D5 and D6 are linked to the Tremolo boost section and intended to help in sound shaping. So it wouldn't matter even if one of them was open since they were in series and not intended as a protection, i.e. not connected to ground on one side and even if it were, the good one would prevent complete malfunction.

These are marked as IN914 diodes. Will check if I can still find the same replacement part at Mouser but would appreciate some guidance on this. I might also try desoldering the open D3 and see what happens. Oh and also... I got my new EDS-88A in-circuit cap analyzer today and did check all the caps including the main power filter caps and they were all either in the green or slight yellow range, no reds! Seems like a neat tool to decide which cap to try replacing - I do know that there is a school of thought that advocates replacing all caps on an old amp while some only say replace ones that are obviously bad and some that might be. Apparently only experience helps determine which ones "might be"! Well this meter might add a little bit of science to the voodoo magic!

Will keep you posted...

p.s. Here is the
IN914 datasheet:

Is this the correct part?