Fender Stage 112 distortion

[redneckgeek]

This amp came to me with a faulty TIP 147 Darlington PA transistor. I replaced all 6 TIP 142'S and 147'S. Now the amp has a distorted sound almost like a bad speaker. If one string is played the sound is clean. If two or more strings are played the sound is distorted.

I have checked all op amps for the presence of DC on the outputs. I have sent a 1kHz signal into the input. The signal looks fine (sine wave) all the way through pin 1 of U7. It is good and clean on the bases of the PA transistors. The presentation on the collectors of the PA transistors in "bell shaped on both the pos and neg side of the trace.

The voltages on the PA transistors are: B=1.10 VDC C=46.7 VDC and the E's=4.8mV. There is a ripple frequency on the 46V collector voltage that measure 100 mA. There is no hum either with or without a signal applied.

I checked the speaker by playing it on two other amps. It is fine. I also played the 112 amp into two different 12 inch cabs and it sounds distorted on them as well.

Looking at the signal at the speaker leads shows this: If one note is played the trace is clean and wavy. If two or more notes are played simultaneously, the package is "jumpy' and seems to break up.

Any input you could give would be greatly appreciated.

[Enzo]

Scope your power rails, either excessive ripple? And the low voltage rails as well.

If you had bad output transistors, make sure to check those low resistance ballast resistors at each one - the 5 watt cement fractional ohm ones. COuld find open ones.

[redneckgeek]

What is Excessive ripple?

I mean, I know what ripple is but what I have never been able to find is some rule of thumb as to what is "excessive"? I also have trouble recognizing what circuitry and or transistor combinations are used to regulate ripple within the circuits it's voltage powers. I have checked the appearance of the voltage on the collectors of the TIP's It was 47V with a ripple of 100 mV pp. I did not check the negative rail but will do that today. I will also remove and test all ballast resistors.

I have Q1 and Q2 on order (2N4401, 2N4403). I was checking the signal on their collectors. The signal on Q2 (PNP) was a half wave but the signal on Q1 (NPN) was three waves of a goofey looking square wave that went flat after three waves. I was also going to replace the zener diodes CR19 and CR27 (15V). I don't see any voltage on either side of them.

Thanks very much for the reply once again.

[Enzo]

It is important to always check both polarities. The power amp runs on +/-47v or whatever, both should be up to voltage and about the same other than polarity. In fact the voltage is less important to me than the same-ness. 46v 48v 50v, just get in the ball park, it will vary as your mains voltage varies anyway. But if I see +47 and -39, then something is off.

RIpple? I just go by feel at this point. 100mv is nothing, count your blessings. At 47v I guess I don;t want to see more than a volt of ripple. What I would be checking for is that say +47 is clean (100mv is clean to me) but -47 has 3 or 4 volts of ripple, something like that.

Ripple isn't regulated per se. In the powr amp, we just rely on those main filter caps to smooth the supplies sufficiently. The voltages for the preamp and the IC in the power amp are regulated, which in a sense leaves the ripple behind. Those 16v supplies come from a couple of zeners. DO you have clean smooth DC +/-16v on the ICs? If so, fine, if not, it may be your problem.

Q1, Q2 are limiters. They work through CR17, CR28. Unsolder one end of those two diodes and lift them from the board. Now Q1, Q2 are disabled and out of the circuit. If that improves the sound, then they may be bad, but if not, l;eave them out for now, the amp doesn;t need them to amplify.

You don;t need to remove the ballasts, just measure across them. They will either be OK or open. They are half an ohm each, which looks about like a dead short to your meter. SO if you get a reading of something more than an ohm, find out why. If they check like shorts then they are probablt fine.

The two zeners you want to replace? If no voltage gets to them, then there will be no voltage across them either. If they are not shorted, and act like diodes, then forget them.

This amp is like most SS amps heavily fed back on itself. SO it is important the feedback is OK. R82 and R83-87 should measure like shorts.

[redneckgeek]

I have been at the Stage 112 most of the day. What I have determined is that I have a case of crossover distortion, and I think the feedback circuits are adding to the problem. I have checked resistors R100 thru R105 (3.3K in emitter circuits), The ballast resistors R106 thru R111. R80 thru R87. All checked good. I have replaced Q1 and Q2 (I just couldn't help myself. I had them and I was in there already. The test you told me about, Enzo, of removing one leg of CR17 and CR28 made no difference).

I don't find anything that remotely allows for any adjustment of bias. I have tested every which way with an ohm meter and all things are equal. I have tested voltages (the AC is tested with Ch2 operating. This amp is being used by a guy who plays only on CH1 so of course the CD shows up for him loud and clear). I really don't know what the voltages should be at the base and or emitters of the TIP transistors??? It's like 1.15v on the base and 0.001 or so on the emitters.

I know that if this is CD I should be looking for a short somewhere but I haven't found it. The 16V plus and minus is at the one end of R92 (+) and R93 (-) but becomes less than a volt on the other.

U7 pin 1 has a great sine wave (1kHz from signal generator) which appears at the bases of the TIP's but the output has the distortion. Pin 2 and 3 of U7 has an exact image of the distorted output as does pin 7. (At pin's 2,3 and 7 things jump around on the scope, [flicker]. I think because of the feedback system's attempt to fix it all).

Oh did I mention? I have replaced all TIP142 and TIP143 transistors twice to no avail. I have also checked the speaker output wires for any trace of DC voltage. I have scanned the entire board under a microscope looking for any breaks or shorts or cold SJ's. I have re-soldered most of the board. I have played the amp into two other speaker cabs. I have replaced most electrolytics.

In other words ... I've done all this and more and I'm really going to be ticked when I find out I plugged a wire into the wrong hole all this time.

(One time I had a bass amp sound like this. It turned out to be a driver transistor that had a broken lead inside of the transistor's case. I only found it when I pulled the transistor out to test it.)

[Enzo]

Your feedback circuit may be obscuring the problem's source, but trust me, you cannot remove it. This is not like a tube amp where you can just disconnect the NFB. Without the feedback system, the amp wouldn;t even function.

A volt or so on the TIP bases sounds about right, and not much on the emitters as well.

Crossover distortion wouldn;t care how many notes you are playing. If you ran a strong sine wave through it, then the output would look like a sine wave, except right where it crosses zero, there would be a small notch - the signal goes horizontal for a brief instant each time it crosses zero. Crossover distortion only involves the part of the waveform that crosses over zero. It doesn't affect the tops of the waveforms.

Your bias is provided by CR13-16. I expect about a half volt drop across each. check for that. If all four of them dead shorted, then you'd have crossover distortion, and if one opened, then yuo'd be blowing outputs. So at about a half volt each, I'd expect to see about a volt at the bottom of R92 - 15v dropped across it. If the bases of the TIPs are at about a volt, then the bias would seem to be working.

Scope the output. Does it make a difference whether there is a load or not?

[redneckgeek]

Thanks for the reply. I got right on it.

When I made the statement about playing two strings I was referring to what I could hear when I did that. The crossover distortion thing came to mind when I used the sig gen. The voltage drop across each of the diodes CR13-16 was 0,5 V each. The voltage at R92 was1.21V (to ground). I scoped the output with no load. It was perfect. with the speaker connected the distortion was present. I am going to try to attach pics of the wave form. Again. Thank you for your input. I really appreciate it. The distorted waveform looks exactly like what you described.

[redneckgeek]

Stage 112 wavesforms

Sorry. Here is a scan of the waveforms.

#725 = output with no load
#726 = pin 1 of U7
#724 = output under load
#727 = pin 7 of U7

Thanks again.

[Enzo]

I see a schematic, I don't see any waveforms.

If a waveform looks good unloaded but failed under load, my first suspects are the outputs not being able to supply the current. In this case, the speaker draws its current through the center tap of the power supply, and through those filter caps. Can you replace the main filters? Or sub in something reasonably close for test?

[km6xz]

Or some of those emitter resistors are open, or transistor pin soldering. The driver obvious is able to drive the no-load output terminals but not getting the current boost needed for a low z load. What do the emitter resistors drop in operation....all the same?
The wave forms would be a big help. When the description of Bell Curve was mentioned, I imagined an output in class B or BC due to level shifting diode array being shorted.

[redneckgeek]

Here are the wave form pics. I have just installed two new filter caps. (3300 uF at 63V). The output wave under load is the same.

As for the emitter resistors. Each one has had one lead de-soldered, tested and replaced. The TIP transistor have been changed twice and each solder joint was inspected under a microscope.

The original complaint on this amp was hum and blowing fuses. I fixed that by installing new TIP 142 and 147 transistors. The distortion has been there since.

I have individually tested the diodes in the PS as well as all resistors. Could it be that the short that caused the original problem also damaged the transformer so that it is not capable of delivering enough current?

The voltage across R100, 101, 102 NPN side is 18.1mV. The drop across the PNP side R103, 104, 105 is 21 mV

[redneckgeek]

I went ahead and removed D113-116 and tested them on an analyzer. They all checked out good

I also tested the windings of the mains transformer. Their Q was high. The resistance between CT's and ends was equal. The 45V side was 0.75Ohms with half that equally on both sides. The 16V winding was 4.8 Ohms with eqaul readings on both sides of the CT.

I re-soldered all TIP 142 and 147 pins.

I tested C49 R75 R76. All were good. As was R68 and R69.

[John G]

Hi,
I have just spent the last few weeks on and off trying to make one of these amps stable. Many TIPs have been roasted, R75-76 fried !!!!
You may want to obtain the Fender Tech notes TN 94-8 and TN 98-1 and its supplement.
This is one horror of an amp with not just one potential problem................................... but many.
Thanks,
John G

[Enzo]

No, it is not the transformer.

You may have tested those emitter resistors, but the real test is to measure resistance directly from the emiiter pin on the transistor down to chassis ground. A resustor might be OK but the traces that connect it could be cracked or burnt open.

DId I ask this? Measure across 10 ohms 1W R82 in the middle of the board next to the coiled wire. It should measure zero ohms, not 10 ohms. This is because it is parallel to the wire coil. If the coil breaks off, the resistor is left alone and wil then measure the 10 ohms.

[redneckgeek]

Oh Boy! Input! Thanks for getting back to me John G and Enzo. I checked the resistance on R82. It was 0.11 Ohms. I measured resitance from the TIPs to ground.

Q3 thru R106 = 0.64Ohms
Q4 thru R107 = 0.62
Q5 thru R108 = 0.60
Q6 thru R111 = 0.62
Q7 thru R110 = 0.61
Q8 thru R109 = 0.61

Also the resistances of R's 100 thru 105 all are right on 3.3K Ohms. They were each removed at one end and measured. The resistance from their common end to ground is 1.1 k Ohms on each side.

Is it possible that the feedback is too weak to result in enough gain to drive the output adequately?