Welcome to Solid State Guitar Amp Forum | DIY Guitar Amplifiers. Please login or sign up.

March 28, 2024, 04:49:39 PM

Login with username, password and session length

Recent Posts

 

Fender RocPro 700 (hard clipping)

Started by Amp1, May 19, 2015, 05:36:42 AM

Previous topic - Next topic

Amp1

Hi guys (and girls?), new member here.
I found this forum very helpfull with polite debate and full of usefull information.
Therefore I'm kindly asking for some help with SS guitar amp.
I try to repair tube amps from time to time, ... this time I have SS amp on the bench.

It is about Fender Roc Pro 700.
Problem is, amp is clipping/cutting off, or farting, on stronger attack.
So, at low volume it sounds nice as it should, but if you turn volume a bit higher and hit the chord it sounds like blocking distortion.
It seems it goes to full conduction (short DC pulses) at peaks only.
It is related to clean and drive channel, but in clean is clearly noticeable.
If I hook guitar (vintage strat) into Return of FXloop I get same problem, less noticeable as this input is less sensitive.
Guitar output is max around 100mVAC (volume and tone stack full open).
If you hit the chord hard, I may turn up volume in clean channel to max 2-3 without hard clipping on peaks.
Measuring signal (scope) at Send output, signal is not clipped.
Also output of driver chip MC1436 (pin 7) is clean, problem is in power amp (TIP142/147), I think.

Testing with tone generator&scope clean/normal channel:
Input singen: 40mVac (volume max)
Output dummy load 8,2ohm: 22,7Vac before clipping
Input singen: 190mVac (volume half)
Output dummy load 8,2ohm: 22,8Vac before clipping

Testing with tone generator&scope on Return input:
Input singen: 140mVac (att.=-7dB)
Output dummy load 8,2ohm: 22,5Vac before clipping

I did change electrolityc caps, all of them.
All solder joints were inspected and mostly re-soldered and clean.
Heatsink and transistors (TIP142/147) were cleaned and fastened with thermal paste.
Power supply DC voltages are all OK, as per schematic
(+42,6Vdc/-42,2Vdc; -17,60Vdc/+17,36Vdc)
DC on output (+/-), no loudspeaker, is around 2mV.
Output transistors are getting warm with no signal on input, but not hot, after 15min.
Voltage across emitter resistors 0,47ohm/5W = around 48mVdc each (100mA), no signal, no load
Bias, emitter-base TIP142=1085mV; TIP147=1113mV; no load, no input
-------------------------------------
Any thoughts regarding biasing output transistors?
Probably, if you look at scope photo, it is kind of oscillation problem?
Photos:
*sin wave after begining of clipping
*square wave before clipping

Also, when you hook a scope to such output – where to put ground of scope?
Minus of speaker is not at the same potential as ground!
(Photos are made without ground of scope connected, otherwise amp oscillate)

Roly

Hi Amp1, welcome.


Quote from: Amp1If I hook guitar (vintage strat) into Return of FXloop I get same problem

So it's in the power amp.

Quote from: Amp1when you hook a scope to such output – where to put ground of scope?
Minus of speaker is not at the same potential as ground!

CRO ground to amp ground.  With the load disconnected there is no signal on the current feedback point "CS2", so we should be able to simply deal with this as a normal voltage amp on "CS1".

Quote from: Amp1scope photo, it is kind of oscillation problem?

Don't think so, the wobbles on the peaks are mains ripple on the supply, normally not an issue, but the flattening off of the waveform coming to the peak is interesting (but without a ground reference...).

The idle bias is set by CR15-CR18.  100mA idle does seem a tad high.

Quote from: Amp1output of driver chip MC1436 (pin 7) is clean, problem is in power amp (TIP142/147), I think.

It looks that way.



With conventional transistors I'd be pulling them for an out-of-circuit test, but I don't have much experience with Darlingtons and their failure modes so I'll wait for the Brains Trust to weigh in on this one.
If you say theory and practice don't agree you haven't applied enough theory.

J M Fahey

Both scope traces look absolutely normal.

Being very finicky, the sinewave is not *perfect*  but as Roly noticed it "kinks" a little before reaching the rail voltages (which are the clipping points by definition, voltage can't go beyond them) at, say, 80% the peak voltage, but that's no big deal and probably normal in that amp type.

I guess you measured with the actual speaker connected.

QuoteProblem is, amp is clipping/cutting off, or farting, on stronger attack.
So, at low volume it sounds nice as it should, but if you turn volume a bit higher and hit the chord it sounds like blocking distortion.
It seems it goes to full conduction (short DC pulses) at peaks only.
Well, the scope shows none of that but plain standard SS amp clipping.
That's what plain SS amps do.
*Very few* (think Pritchard, Quilter, Peavey Transtube, a couple others) doctor the amps so they clip different, more "tubey" if you wish, 99% out there do not.
SS amps are usually very clean when clean, more than tube amps, and "bump" into clipping with no warning in a somewhat harsh way, I guess that's what you are hearing.
I suggest you connect the scope to output, with speakers connected, and play the guitar; I guess the harsh sound you do not like matches visible signal clipping.

You won't have a stable, in sync image because guitar chords are a mix of frequencies, but you'll clearly see the tops flattening and the trace getting thicker/brighter there .

If so, your amp is fine.

You might play a few loud chords and have somebody else take a screen picture to post here.

g1

  Can you verify that it is not some vibration caused by the high volume that is triggering the problem?
Running guitar into FX return, if you give the cab a good solid thump (with palm or rubber mallet or what have you), does it affect the symptom?

Otherwise, maybe others here can comment whether the 100mA bias sounds correct.

Amp1

Thank you guys!

@Roly
Yes, I also think it's power amp, I can't believe it's "normal" or input is too sensitive.
Thanks for tip on scope ground, video is recorded this way now.
Idle Bias is set via diodes I thought, yes.
Wouldn't be better to set it a bit colder? Any sugestions on best and simplest way?

@g1
For sure I will try your tip, but I have to put chasis back to combo.

@J M Fahey
I don't know if it's normal, usable volume on clean channel is up to 2 of 10. (Guitar vol max)
If I set guitar vol 1/2 and amp vol 1/2 is almost good enough, but it's not 65W then, much less.
Yes, photos of scope traces are made with dummy load hooked on output.
Your explanation of plain SS amps scary me that this one it's like it should be :(
I didn't have much SS amps on bench, but this one surprised me (comparing to tube).

I recorder a clip, it's on youtube, hopefully you can see/hear what I'm talking about.
Amp is sitting on bench, chasis is out of combo, speaker is EVM12L (covered with towel).
Begining: amp1/guitar10 - then amp5/guitar5 - then amp10/guitar3...
https://www.youtube.com/watch?v=9q7VtuyCtzA




Roly

{ogg! One of your strings is flat :-\ }

Well something is clipping alright.  The CRO Y vernier looks like its uncal so I can't tell what the actual peak to peak voltage is.  Swing it around to "cal" (far right) and see if you can measure the actual peak-to-peak voltage you are getting when it clips.  A sine source is much better for this; you could try recording a 10 sec 500Hz .WAV in Audacity or similar.  There are also test tone downloads.

If the p2p at clip is around 80V, rail to rail, then what you see is what you get and you have discovered what some of us call "the wall" with s.s. amps - you just avoid OP stage clipping 'coz it sounds nasty.

On the other hand if the clipping is occurring at a much lesser voltage that might give us a clue to a problem.

While we don't yet know what the actual p2p output voltage is at clip, I have to agree with JMF that that looks and sounds pretty normal.  {head scratch}

Have you checked it patched to a different cab to eliminate a speaker fault?  Frayed speaker braids can give some very nasty intermittent farts.
If you say theory and practice don't agree you haven't applied enough theory.

phatt

The speaker sounds like it's getting too much bass or it's half shot.

It's possible someone may have already tried to modify the circuit, the fact it has a non standard speaker is a clue and there maybe other mods inside as well.

I have had experience with a Rock pro 1000 which was rather bottom heavy as stock.
Even with the stock circuit and speaker you had to go easy on the bass knob if you wanted to play very loud.

Maybe back off the bass knob and see if it can play clean and loud without clipping. If you can do that then the system as a whole has too much bottom end and may need tweaking.

Your issue may not be unique as a lot of old Valve Amps farted with high volume and hi bass levels.

Meantime take Roly's advice,  plug in a known good speaker and check that first.

Phil.

J M Fahey

Agree with above posters.

Sound is WAY too bassy, zero mids and treble, how are you setting amp EQ?
What is your guitar?  It might have very hot Metal pickups.
How is guitar tone control set? Bassy/flat?

Where is that EV speaker? Bare on the table? In a cabinet somewhere else? What does towel covered mean?

Waveforms look normal, clearly showing top and bottom peaks clipping, and show heavy treble cut.
Bottom peak seems to clip earlier than top one, but guitar signals are unsymmetrical, so that might explain it.

Feed 100mV 1kHz signal at the input, set all tone controls to 10, Master also to 10 if available, go through the clean channel and sssssloooowwwwlllllyyyyyy rise volume until signal just clips , always with some load, speaker or 8 ohms resistor.

Measure RMS value of signal and calculate power: V*V/8 = ??
I expect around 20V RMS ; double verify by measuring Vpp on screen, should be around 60/65Vpp (there's always some loss across transistors).

SS amps do not clip gracefully (that's an understatement)  but still are very usable, proof is that 100 SS amps are sold for every tubed one.

Clever EQ, pulling "ugly" frequencies away is a great trick to make them a practical option, Teemuk in his book (available here for download) has a chapter about that.

Plus 99% pedals are SS, yet many of them sound very good, also thanks to clever EQ, both pre and post distortion; our friend Phatt has also experimented a lot and even designed an add-on EQ (PhAbbTone)  which improves most amps.

Here's a demo of a Roc Pro 700, unmodded and into its original average quality speaker:
https://youtu.be/_rP58IAV4ko?t=3m39s

Retest and post results.

Almost forgot, download test tones from here http://www.mediacollege.com/audio/tone/download/
Any cheap MP3 player will give you around 200mV RMS, full blast into the headphone out, pad down as needed.

Amp1

#8
I've been busy with other things, now I'm back.
Thank you for all your sugestions and opinions, always welcome!

I think I found a bug...pretty fast this morning.
Since preamp is fine and everything checked I fosuced into power stage including driver&Return input.
When I was measuring voltages (during hard clipping, with speaker connected) around driver chip I heard,
that clipping/oscillating stops if I touch specific points.
These points include pin6 of MC1436, both bases of TIP142/147, around all biasing diodes.
Feeding sin signal into return I got clipped (oscillation) positive top of wave - see photos.
It begins at specific (low) level of input signal, no matter of frequency.
Photos: upper signal is pin 6 of MC1436 and lower signal is from speaker.
One photo is made before adding ceramic cap and one (no clipping) after.

So I soldered a cer cap 47pF across base -collector of TIP142.
Oscillating is not present anymore, I got clean 22,5Vac on 8ohm load, as it should be.
I added trimpot 250kohm instead of resistors R14 and R15, to be able to further regulate/decrease level of signal through preamp.
Input was very sensitive, too much.
Guy using this amp is folk guitarist with Framus AZ10 and low output Zoller pickups, he needs clean only.
Now we are both satisfied  :), sound is really nice, Fender (SS) clean and usable power.

Regarding sound from clips.
I recorded it with iPad, close to speaker, so I put a towel accros it.
It was too loud to be recorded usable and I was just tried to present that clipping sound...
Speaker is 8ohm EVM12L in HQ 18mm birch box, made excatly following EV plan TL806, very good experience!
Although these speakers sounded very good on cleans, they're not as good for overdriven signals, my opinion.

As test signal sin&sq I use LAG26 (0,1% dist) ton generator and HP8903 distortion measurement system (sin only)
For  RMS AC signals there is Fluke 189, Marconi TF2337A (RMS 1kHz&400Hz and also AC part of HP8903.






Roly

Ooow-arr!  That there's a parasitic oscillation all right, now we can see it.

Looks like you nailed it - well done.   :dbtu:

{Now my only question is, why would a modern amp from a reputable builder have such an obvious parasitic (if nothing else is wrong)?}
If you say theory and practice don't agree you haven't applied enough theory.

Amp1

Thank you Roly!
Well, it was done with a little help of my "ssguitar" new friends ;)

You said "If nothing else is wrong"...
...yep, something for sure is wrong if I need to add a cap to stop oscillating.
I'm thinking about several reasons, but I'd let specialists to begin with this topic if they wish.
Power supply is stable, DC voltage drop is minimal on full power.

Probably it would be better (and faster) to change all elements in power amp.
This way you dissmount and assemble everything once only.
I hate soldering/desoldering old PCB, it's always dangerous to destroy cooper traces.

J M Fahey

Congratulations on solving it  :dbtu:

As of:
QuoteProbably it would be better (and faster) to change all elements in power amp.
ABSOLUTELY FORGET that.
You will add a dozen new problems and solve nothing.
You will subject the PCB to hundreds of desolder/pull/resolder operations (1 per hole) , with countless occasions for mistakes, and even if you do it right, new parts will be different from original ones (duh!!)  , *all*  electronics parts have tolerances, and the new ones may add up the wrong way and you end up with a far worse amp.

In fact I think something similar may have happened :
Quote
{Now my only question is, why would a modern amp from a reputable builder have such an obvious parasitic (if nothing else is wrong)?}
You are generalizing from a sample size of >1< , versus tens of thousands Fender amps running happily all over the world with exactly the same power amp schematic.

Transistors vary in their characteristics and a common problem when repairing old stuff (I avoid calling "vintage" stuff made in the 70s and 80s ;) ), specially in Audio/HiFi stuff which has larger wideband than guitar amps, is that new devices, even with the same code, have better frequency response (good)  but now older compensation does not work, or works marginally.

Something may have been replaced and compensation became marginal, or something similar.
The fact that just 47pF were enough makes me think that thatn particular TIP was somewhat faster than normal, or maybe the Zobel network was open ...

Enzo

it is tempting to just replace all the parts, but remember, we are dealing with a circuit, not parts.  What if the problem is a broken or cracked printed circuit trace?  You change every part on it, but the crack remains.  That is why we want to always identify the actual problem and then fix that.

Roly

Quite agree about the futility of a total rebuild, but I think (hope) that this was a throw-away line rather than a serious suggestion.

Quote from: J M FaheyYou are generalizing from a sample size of >1<

No.

Quote from: Rolywhy would a modern amp from a reputable builder have such an obvious parasitic (if nothing else is wrong)?

I.e. I don't think this is generic to this model - it is because I don't think someone like Fender are knocking out marginally stable amplifiers that I posed the question.

What I'm wondering is if this newly introduced rolloff is only needed because the amp has lost a rolloff or HF bypass somewhere.

Quote from: J M FaheyTransistors vary in their characteristics

Which in turn invites the question, didn't Fender designers fully account for possible device spread?

I rather doubt that (given that they would have done a Monte Carlo and a min/min max/max on the design).  Given that these have been built in large numbers I would expect such a problem to emerge from production line testing - an accumulating pile of problem units in the corner of the factory.  {been there, done that  ::) }

Quote from: J M FaheySomething may have been replaced and compensation became marginal, or something similar.
...
or maybe the Zobel network was open ...

Exactly my own thoughts.  I'm suspicious that the other boot has yet to hit the floor.

Again I stress that, until evidence turns up to the contrary, I don't think this is a model generic problem, rather something specific to this particular amp has been overlooked.

Enzo goes to the heart of the matter;

Quote from: EnzoThat is why we want to always identify the actual problem and then fix that.

Is this an actual repair, or is it only wallpaper covering the actual fault?  This particular amp seems to be fixed, but is it actually repaired?  That is my residual concern.

As you say JMF, this amp has been built in large numbers, and if it's a generic problem of component spread then there should be other examples of parasitic instability in this model.  ATM we have no evidence that this is the case, so the obvious conclusion is that there is still something not quite right about this particular unit.

Probably too late now to follow up, but I have a feeling we know the what but not the why.
If you say theory and practice don't agree you haven't applied enough theory.

Amp1

I agree, the source of problem is not found, it's still there.
That 47pF cap is just a pill, pushing the pain back to acceptable level...
Also agree, with loss of knowledge&practice it's time consuming procedure - to find a reason.
Such taking pills I found as common practice of repairman... 8)

Interesting/similar case with Fender Proreverb (all tube amp):
The same problem occured on friends Proreverb (I think 74').
Amp put out farting noise (audible, but not as hard as from Roc pro), after some level of volume.
I noticed same oscillation scope form on bias rail.
When I took bias voltage from my clean bench source, problem dissapear...
There were several revisions of Proreverb schematics (slightly changed over time).
Comparing all those, the amp I had on the bench was without small ceramic caps from bias (tube pin) to ground.
After I added those caps (according to newer revisions of schematics), oscillations gone, were not visible and audible anymore.