Princeton amp AA964 - PT rewound - Rob robinette mods - readings and behavior

[spunko]

This is the amp I mentioned in another thread, it was a total mess. Now every single wire has been replaced and the power tranformer has been rewound.

The power transformer now gives 330VAC instead of 340VAC shown on the schematic, so all the voltages are low vs factory specs. The amp sounds good tho.
Also the PT gets hot, I can leave my hand on it all day, it won't hurt but still hot, not warm. This is my first princeton so I'm not familiar with it.

I also did the suggested mods by Rob Robinette: Adjustable bias (already in the amp), phase inverter grid stop, power tube grid stop and screen resistors. Everything works,

BUT!!  The pilot light oscillates with the tremolo effect  The owner said this behavior was present in the amp since he got it, so it followed the rebuild.

This is what I noticed. With the footswitch plugged in:
- Tremolo OFF - Speed '0 - Intensity '10 -
Current draw on 6v6s goes up by 2mA on both tubes from 22 mA to 24 mA (this time I biased the amp a bit hotter). No pilot ligth oscillation.

- Tremolo ON - Speed '0 - Intensity '10 -
Current draw on 6v6s oscillate. Pilot ligth oscillates.

What could cause this symptom? The pilot light is a separate circuit from the bias. But they share the same ground point in the chasis with the virtual center tap from 6.3v heaters.

The readings in RED were taken with cold bias, max negative voltage. Power tubes were drawing 16.4 mA both.

[Latole]

PT look under power ; what are the heater voltage ? Under 6 AC it is the PT

Bias at 16.4 is  50% power dissipation , I never biased a amp more the 60%
Often at 50% if sound is good. 70% is too high IMO

[mresistor]

Have you substituted known good 12Ax7 tubes   one at a time to see if the problem persists? 


is there any solder in this eyelet 

[Latole]

Have you substituted known good 12Ax7 tubes   one at a time to see if the problem persists? 


is there any solder in this eyelet

[mresistor]

Still hard to tell   

[uki]

The pilot light oscillates with the tremolo effect 

That is very interesting!!! To me it is a great feature !! Very cool !! as long it doesn't interfere with the amp functioning, why not !!

[spunko]

PT look under power ; what are the heater voltage ? Under 6 AC it is the PT

Bias at 16.4 is  50% power dissipation , I never biased a amp more the 60%
Often at 50% if sound is good. 70% is too high IMO

Forgot to mention, PT gives 6.3 VAC for heaters (re-checked). The only low voltage is the 330VAC.
When the pilot light oscillates with the tremolo, the 6.3 VAC reading does not vary that much. My digital meter shows between 6.39 VAC to 6.31 VAC.

The problem is the amp behaved like this since the owner got it years ago, before smoking the PT, before rewinding the PT. Owner always plugged in the footswich to "counter" it.
When he mentioned it, I thought maybe was a grounding issue, or all those corroded wires, but the symptom persists.

Have you substituted known good 12Ax7 tubes   one at a time to see if the problem persists? 

is there any solder in this eyelet

Yes it is soldered, just re-checked, 99% of soldering was done from bellow the eyelet board.

Yes I put two new 12ax7 with the same result.

That is very interesting!!! To me it is a great feature !! Very cool !! as long it doesn't interfere with the amp functioning, why not !!

I like it too, but I want to be safe and also understand why is this happening.

[AlNewman]

You'd have to wonder if one of the old brown caps tied into, and/or separating the trem circuit from the bias may be leaky.

[stratomaster]

Some transformers that fender used have the heater CT internally grounded.  Adding an artificial CT to these transformers can cause problems. It's possible yours is one of those types.

[Latole]

"When the pilot light oscillates with the tremolo, the 6.3 VAC reading does not vary that much. My digital meter shows between 6.39 VAC to 6.31 VAC. "
- spunko

It may be not good.
And a digitals meters are very slow , Voltages reading may vary more
And PT get too hot

1+1 = 2 .  If there are no short in the circuit, PT is underpowered

[spunko]

"When the pilot light oscillates with the tremolo, the 6.3 VAC reading does not vary that much. My digital meter shows between 6.39 VAC to 6.31 VAC. "
- spunko

It may be not good.
And a digitals meters are very slow , Voltages reading may vary more
And PT get too hot

1+1 = 2 .  If there are no short in the circuit, PT is underpowered

I just got a '79 vibro champ PT, same model and this one is printine, also the amp is flawless. Swapped PTs and still the pilot light oscillates with the tremolo but I think is less visible. Is really sunny today so I may test at night again. It also gets hot like the rewound one, but it won't hurt.

I took some readings to compare both vintage PTs (pic).
Mojotone modern replacement shows 320V, hammond 325v, are these new PTs an upgrade in terms of manufacturing/materials? These show lower voltage.

[AlNewman]

I would check DC voltage across the 1M resistor tied to the plate of the trem tube.  There should be 0 voltage drop.  That would tell you if one of those caps is leaky.

[uki]

Check this out, this video describe your pilot bulb effect, how it happen

=91

[SEL49]

You currently have the bias cap grounded at the same place the 100Ω artificial CT resistors are grounded. Move the bias cap ground wire and the RED/YELLOW PT wire to the cap can lug. See pic. Does this help the tremolo pilot lamp?

Also, just curious. I noticed you have a non-standard green bezel lens on the pilot lamp. Verify the lamp bulb is a standard #47, maybe even try a new bulb.

[spunko]

I would check DC voltage across the 1M resistor tied to the plate of the trem tube.  There should be 0 voltage drop.  That would tell you if one of those caps is leaky.

If I got it right there shouldn't be voltage on the left leg of the Speed pot, but it reads 1.5VDC.
Also, I have to plug in the FS or short the trem jack to take the reading. If I don't shut off the trem effect, I can't measure the voltage on pin 1 of the trem tube because it varies or oscillates with the effect.

I took the PT back to the rewind shop, they'll do it again to get the correct voltages.
While they fix it I'll do more testing with the vibro champ PT. It gives much higher voltage, 442 VDC on the 6v6 plates.

The problem is, the pilot light oscillates with both transformers. 

You currently have the bias cap grounded at the same place the 100Ω artificial CT resistors are grounded. Move the bias cap ground wire and the RED/YELLOW PT wire to the cap can lug. See pic. Does this help the tremolo pilot lamp?

Also, just curious. I noticed you have a non-standard green bezel lens on the pilot lamp. Verify the lamp bulb is a standard #47, maybe even try a new bulb.

The ground point is factory, but I will move it.
Yes the original jewel is long gone, the lamp jack is original tho. Also new bulb because the old one blew with the PT when it smoked.

[SEL49]

The ground point is factory, but I will move it.
Yes the original jewel is long gone, the lamp jack is original tho. Also new bulb because the old one blew with the PT when it smoked.

I don't think Fender used solder lugs connected to PT bolts for a ground point. When that amp was new that red/yellow wire was soldered directly to the chassis between the PT and the cap can. All that rust may have created a poor ground for the bias cap and 100Ω resistors and that 'may' be the reason for the odd trem/pilot lamp behavior. Connecting the red/yellow wire and bias cap ground to the cap can ground lug will provide a solid power ground for the amp.

Is the lamp bulb actually a #47?

[Lauri]

I don't understand why anyone would consider the pilot light brightness changing with the tremolo a problem. It's completely normal behavior in amps with power tube bias wiggle tremolo.

[Willabe]

I don't understand why anyone would consider the pilot light brightness changing with the tremolo a problem. It's completely normal behavior in amps with power tube bias wiggle tremolo.

Ok, and your basing this on what?

I don't recall ever seeing anyone post here asking about this before. And there have been many bias wiggle amp clones built here. I built a 5G9 Tremoluxe and it doesn't do that. 

[spunko]

Is the lamp bulb actually a #47?

It is a #44, I know I have a #47 in a drawer. It will have to wait because I took the OT to the shop to replace the wires, they are stiff as hell, the OT is good tho.

[SEL49]

#44 is a higher wattage bulb than #47. Using a #47 bulb and moving those grounds just may solve that weird issue.

[AlNewman]

If I got it right there shouldn't be voltage on the left leg of the Speed pot, but it reads 1.5VDC.
Also, I have to plug in the FS or short the trem jack to take the reading. If I don't shut off the trem effect, I can't measure the voltage on pin 1 of the trem tube because it varies or oscillates with the effect.

I took the PT back to the rewind shop, they'll do it again to get the correct voltages.
While they fix it I'll do more testing with the vibro champ PT. It gives much higher voltage, 442 VDC on the 6v6 plates.

The problem is, the pilot light oscillates with both transformers. 

You're right, there should be 0 DC voltage (or thereabouts) at the top of the speed pot.  If that was the grid of a tube, the caps would need to be replaced.  There's also capacitors from the plate of the trem circuit that could directly influence the DC bias of the power tubes.