Fender Princeton '65 Bias (?) Problem

[Tom Phillips]

OK. Then we are left with more detailed troubleshooting which you indicate is out of your ability/comfort zone or g-one's suggestion which is also probably out of your comfort zone. Correct?

[John R Smith]

Tom

I have done quite a bit of messing about with my stereo valve amps but that was all with point-to-point wiring, no PCB boards. That circuit board looks very intimidating, and you don't have nice long tails on all the components to make soldering easy. I have always stuck to my belief that a little knowledge is dangerous, and I know full well that a pro would probably have this problem licked and out the door in 30 minutes where I could easily make things worse through my own lack of expertise. So that is why I am a bit hesitant to pursue things further. For example, I find the value of R28 to be 100k rather than 120k as per the schematic, but I don't know if that is just poor measurement on my part, or whether it matters, and I'm not sure that a lower value there would not increase the bias voltage rather than decrease it anyhow. See what I mean? Fools rush in . . . .

[Jazz P Bass]

I find the value of R28 to be 100k rather than 120k as per the schematic, but I don't know if that is just poor measurement on my part, or whether it matters, and I'm not sure that a lower value there would not increase the bias voltage rather than decrease it anyhow.

One way to find out is to measure TP19.
It either is -40Vdc or it isn't.
While you are on it, flip your meter to read Vac.
That will give you a measurement that will tell you if the supply is stable.

As for the ground reference for your voltage measurement, use the 'striped" end of D4 for your probe connection.

[John R Smith]

Oh well, my curiosity got the better of me. So this afternoon I checked and found that the battery in my meter was low so my resistor readings were off. Sorted that and checked again -

* Bias voltage at TP19 is -31V, not enough as before

* Tested R2 is correct at 27k

* Tested R28 is correct at 120k

* Tested R22 bias pot is correct at 10k

* Disconnected one leg of D4 as suggested, no change to the bias voltage at TP19, still -31V

So I think that is about all I can test around the bias circuit without pulling the board. Any other ideas?

[52 Bill]

You need to find out if the problem is that there is not enough bias voltage being generated or if the circuitry is pulling the voltage down. Measure the dc voltage on the supply side of R28, that should give you an indication of what's happening in the circuit.

I noticed that all of the dc voltages are a little low, is this amp original to your country or is it being transformer boosted to the correct line voltage?

[John R Smith]

Hello 52 Bill. The voltage on the supply side of R28 is 190V, but there is no indication on the schematic of what it should be, unless I have missed something. The transformer is a 0075167000 which does not fit with any ID on the schematic but I assume is correct for 240VAC.

[Tom Phillips]

... Also disconnect the bias supply from the Vibrato pot (no power tubes in!) and check for proper operation...

John,
Did you ever do that? This would completely isolate the bias supply for further troubleshooting and, if the bias supply voltage returns to normal, then we would know that something in the other circuitry was loading it down.

You can do this by disconnecting the wire connection shown as P18A-3 from the main PCB to P18B-3 on the front panel control PCB.

Note that your meter is going to be confused making an AC reading on the supply side of R28 because the voltage there is a half wave rectified sine wave.
Tom

[John R Smith]

John,
Did you ever do that? This would completely isolate the bias supply for further troubleshooting and, if the bias supply voltage returns to normal, then we would know that something in the other circuitry was loading it down.

You can do this by disconnecting the wire connection shown as P18A-3 from the main PCB to P18B-3 on the front panel control PCB.

Tom

Tom, no I did not do that yet because I didn't know how. So all I have to do is unplug the ribbon connector P18A from the main board, OK that will be the next step. Will report back later . . .

[John R Smith]

Er, right . . .

Before I carried out the test with P18A unhitched I powered the amp up just to do a quick control check of the values at TP19 etc for comparison purposes. It was just as well that I did, because this morning we have -45V at TP19 and TP13 without making any changes at all from yesterday. Rubbing my eyes with disbelief, I powered down and installed the power tubes V5 and V6. After bringing the amp back up and letting it warm through, I have the following readings -

* TP13 and TP19 are both -45V

* TP16 is 28mv, TP17 is 22.5mv

* This is with R22 at maximum bias

* And plate B+ is now up to 462V

So in other words the amp is now running just fine. This does not look good because I must have an intermittent fault which will be a pain to track down. What component could go down and then come back up again all by itself?

[Yelly]

A leaking bias cap C25 could cause voltagedrop. Wouldn't be the first time in history.

[Tom Phillips]

...So in other words the amp is now running just fine. This does not look good because I must have an intermittent fault which will be a pain to track down. What component could go down and then come back up again all by itself?

Well that's bad but also good news because we now have a new clue. The cause could be a component but could also be a solder connection or internal plug/cable connection. One approach to troubleshooting these situations is to monitor the voltage while you tap the various connections & components with a non-conductive prod. This is the "chopstick testing" you ofter hear people talk about. Heating and cooling cycles can also make the fault stand out.

Many techs just start re-flowing solder joints. I like to actually identify the faulty joint first and a good visual inspection under a magnifier often does the trick. However, this requires that you do some dis-assembly to expose the back of the PC boards.

One test that you could do right now is to set up the amp to monitor the bias voltage with the power tubes removed. Monitor the bias voltage at pin 5 of one of the power tube sockets so the whole circuit path is included. The power tubes are removed so you can leave the amp on and come back periodically to check the voltage while you prod around as described above. This way, if the amp goes into fault mode while you aren't watching, it won't stress the power tubes. After you have this set up it would be interesting to know what the min & max bias voltage readings are as you sweep the bias pot through its complete range when the circuit is working correctly. You only had a range of a couple of volts when the amp was in fault mode. I'd expect an adjustment over a range of 15 to 20 V in properly working mode.

This has turned into quite a troubleshooting adventure. Hopefully, you are having a little fun with the learning process. In the end, the cause will be traced to one simple little thing. Sometimes that thing is found during the first 5 minutes and sometimes it takes a long time to locate.

Cheers,
Tom

[John R Smith]

Well, as I suspected the good times did not last very long. I kept the amp powered up for about 1 1/2 hours, plugged my guitar in and played for a while and then - oops, a very hot smell and no tremolo. Back to low bias and glowing plates. So I have an intermittent fault which can somehow randomly cure itself and then cut back in again without obvious cause.

What should be a clue is that when the bias goes bad it is always to exactly -31V, never more or less. Almost as if a component is switching itself in and out. So I then carried out the test as Tom suggests, unplugged P18A and tested TP19 at the bias pot. Bingo, -45V, so the problem is not in the bias circuit but in the trem circuit. Now there seems to be an awful lot of components in there.

Where do I start?

P.S. Sorry Tom, I missed your post above when writing this. I have tried tapping everything in sight with a plastic rod while monitoring TP19, but no observable change in voltage.

[Tom Phillips]

...So I then carried out the test as Tom suggests, unplugged P18A and tested TP19 at the bias pot. Bingo, -45V, so the problem is not in the bias circuit but in the trem circuit. Now there seems to be an awful lot of components in there...

One rule of troubleshooting is the re-verify your finding before saying Bingo to verify that it's not a coincidence. In this case that would be to plug P18A back in and verify that the problem returns. If you verify that plugged in is bad then unplugged is immediately good then plugged back in is immediately bad then you have "Bingo."

...Where do I start?...

With the amp off and P18A-3 unplugged measure the DC resistance from P18B-3 to chassis ground. Do you get "OL" ie. ~infinity? If not then you look for partial shorts or bad components. However, the intermittent nature of the problem will probably cause uncertainties. More poking / waiting / heat cycles may be needed. If it were on my bench I'd be doing the under board inspection too.

[John R Smith]

One rule of troubleshooting is the re-verify your finding before saying Bingo to verify that it's not a coincidence. In this case that would be to plug P18A back in and verify that the problem returns. If you verify that plugged in is bad then unplugged is immediately good then plugged back in is immediately bad then you have "Bingo."

With the amp off and P18A-3 unplugged measure the DC resistance from P18B-3 to chassis ground. Do you get "OL" ie. ~infinity? If not then you look for partial shorts or bad components. However, the intermittent nature of the problem will probably cause uncertainties. More poking / waiting / heat cycles may be needed. If it were on my bench I'd be doing the under board inspection too.

Tom - definitely "bingo" on the P18 plugged/unplugged. I just tried it several times, and at R22 plugged in is -31V and unplugged is -45V. Resistance from P18B-3 to ground is infinite. Sorry I keep cross-posting, I am not used to this threaded software!

[Tom Phillips]

John,
This is good progress in that the bias supply itself seems to be eliminated as the cause of the problem. P18 is a 6 pin connector so when you unplug it, several connections ore opened which include the tremolo circuit AND the bias supply to the power amp section. Next step is to isolate the problem further to determine which path is loading down the supply. Do you have jumper leads such that you could connect P18A-3 directly to P18A-5? This would be done with the ribbon cable disconnected from the main board and it would connect the bias supply to the power amp while leaving the tremolo circuit disconnected. The jumper would be installed at the connector header on the main board. Caution: You need to be really careful that something doesn't slip up and cause a short and resultant further damage.
Tom