Peavey TNT100 (1974 version) transistor bias issue

[telecaster]

Schematic attached:peavey_tnt_100_1974.pdf
History: Had a shorted output transistor, installed a new pair of 2n3055's. Distorted sound - found that 4 out of 5 electrolytic coupling caps (2uf@35V) were installed backwards and leaking DC. Replaced all 5 and the all power supply filter caps while I was in there. Sounded better, but I still had a fuzzy overtone, aggravated by lower bass notes while playing a bass. Found that transistors 1 and 2 (I numbered from left to right on schematic). were replaced with non-original transistors. Coupled with that, the bias voltages for transistor 1 looked wrong (measurements below). I had a 2N3391A for transistor 1 and used an NTE159 for transistor 2 so I replaced them both. Bias voltages changed, but still look wrong because transistor 1 is not forward biased.
Measurements:
With original (incorrect) transistors installed
1. C-20.8 B-13.94 E-14.07 Close to schematic reference voltages except for base about 0.6 low
2. C-13.7 B-20.8 E-21.7
I replaced the 1M and 470K resistors that form the bias for transistor 1's base, no difference in voltage measurements

With 2N3391A and NTE159 installed
1. C-19.4 B-18.8 E-18.9 Again not forward biased
2. C-18.6 B-19.3 E-20.1
Sound did not improve. FWIW all other transistors bias voltages matched the schematic reference voltages before and after changing the 2 transistors. Power supply voltages look correct and all bias resistors measure correctly.

Can anyone help me figure out why the bias voltage is off on transistor 1? Any other help/suggestions would be appreciated

[TomCarlos]

You need to be careful with too much soldering (removing parts, reinstalling parts) etc.... BUT.. if it were me, I would pull the two transistors. I would then measure at the voltage point where the 1 Meg and 470K meet (to ground). That is a simple voltage divider. I would then check the transistors (if you do not have a transistor checker, you'll need to use your ohm meter for this). I know that many of the fellow MEF members are not keen on NTE parts - neither am I. You should be able to find 2N4249 PNP transistors.

[Enzo]

Some random thoughts.

Those little 2uf caps have not aged well, and just about all of them need replacement anyway.

HAve we verified that the speaker voice coil is not rubbing?

2N5087 should work in place of 2N4249.

I don't like NTE parts, but they should function here. MAke sure the NTE leg arrangement is the same as the old part, so the wires are in the appropriate holes.

HAve we verified that the first stage is WHERE the distortion is happening? I mean as opposed to just inferring it due to voltages we dislike?

[telecaster]

TomCarlos, I took your advice - with the two transistors out, the base voltage on transistor 1 was 15.87 volts. I then noticed someone was in here before and swapped the 1M and 470K resistors. So 1M/1470K x 24 V is around 16V. I swapped the two resistors and measured 7.5V (470K/1470K x 24V = 8V). Now I'm confused, if the target base voltage on the first transistor is 14.6V, then the resistors should be swapped vs the schematic.

[telecaster]

Enzo,

Those little 2uf caps have not aged well, and just about all of them need replacement anyway. Yes, I replaced all 5 of the 2uf@35V electrolytic coupling caps

HAve we verified that the speaker voice coil is not rubbing? Yes, I tested using my bench speaker with the same results as the Peavey internal speaker

2N5087 should work in place of 2N4249. Thanks, I actually have some new 2N5087s

I don't like NTE parts, but they should function here. MAke sure the NTE leg arrangement is the same as the old part, so the wires are in the appropriate holes.

HAve we verified that the first stage is WHERE the distortion is happening? I mean as opposed to just inferring it due to voltages we dislike? Good question, I am NOT sure the first stage is the culprit, I will get known good transistors and scope through the amp.

[TomCarlos]

Here is how to calculate the voltage divider for each resistor:

Across 1M: (22 volts x 1,000,000)/1,470,000 = 14.96v

Across 470K: (22 volts x 470,000)/1,470,000 = 7.03v

Total = 22 volts

My guess is that the 1M should go from the Q1 Base to ground. It looks like the schematic is incorrect. Anyone else want to second that?

[telecaster]

Update with scope pictures

Transistor 1 base bias resistors put back as found - 1M to ground, 470K to 22V. Transistor 1 replaced with tested good 2N3391A, transistor 2 is tested good NTE part. Transistor 1 base-emitter bias still off as originally measured, but I went ahead and scoped through the amp to see where distortion occurs to answer Enzo's question. Turns out that the transistor 1-2 pair passes signal cleanly without distortion. The applied signal was 30mv 100Hz, which is where the notes sounded worst when I played a bass through the amp. Volume was about halfway up, tone controls all in the middle, and master volume a bit less than halfway up into an 8ohm load.

Transistors numbered left to right in the schematic attached above:
Picture 1 (leftmost): Transistor 5 collector
Picture 2 : Transistor 6 base
Picture 3 : Transistor 6 collector
Picture 4 : Transistor 7 (driver)collector
Picture 5 : Speaker output (+)

[The Dude]

......MAke sure the NTE leg arrangement is the same as the old part, so the wires are in the appropriate holes......

^^^^^^ That. Did you double check the basing diagrams? If someone put transistors in the amp that were not direct subs, they may have swapped transistor orientation to make them work. The 2N3391A is ECB and the NTE159 is EBC.

[Enzo]

OK, this is just my intuition, my spider sense, so for what it is worth... What I see is the final picture. Looks like it wants to make a sine wave but can't move the negative side. The driver transistor wants to make a sine wave, but the NFB tells it about the missing side. SO the driver TRIES to send a larger signal to compensate.

So first, is that NFB cap, one of your new 2uf guys, installed right, not backwards? I will assume the bottom output transistor is Q9 in your plan? Does Q9 emitter get -35v?

Q8, Q9 are the same circuit, just one bottoms out to -35v and the other bottoms out to speaker -- zero DCV at rest. But you can still compare E-C voltages and such. DO so.

My guts are itching that MAYBE the lower transformer winding is open or broken wire. Schemo says yellow and blue wires.

[TomCarlos]

Follow up to Dude's entry #8 - see photo for the 2N4249 and NTE159 pin out.

[telecaster]

Problem found and fixed, thanks to you guys!

1. TomCarlos and The Dude - I did get the basing diagrams figured out when I was testing replacement transistors 1 and 2 prior to final installation. In case anyone else works on one of these - the original SPS953 transistor 1 was EBC and the recommended replacement 2N3391A was ECB.

2. Enzo - your gut was right on the money! I had originally probed the windings on the transformer which showed continuity on both, as did the primary, BUT I did not verify the connection to/through the board. Upon closer inspection today, one wire on the lower winding had broken just above the board - it was not obvious until I pushed firmly on each wire. Repaired the connection and the amp is working fine.

My thanks to all - I appreciate the help, and nudges in the right direction