Problem with biasing Twin reverb type amp

General discussion area for tube amps.

Moderators: pompeiisneaks, Colossal

tedsorvino
Posts: 12
Joined: Tue Aug 08, 2017 9:13 pm

Problem with biasing Twin reverb type amp

Post by tedsorvino »

Hello kind people of the forum. It's the first time I'm writing.

I've got a mid 70s Acousta amplifier (silverface Twin reverb replica). I'm trying to blackface it. Most things go well. bThe rectifier is a bridge rectifier because of the Power Transformer. It hasn't got a normal channel. There was an active parametric eq (much like super twin) that made things bad, so I took it out.
Now I'm triyng to change the bias to properly adjust my tube with my tube probe.
There is a problem here. The readings were way too low (around 10- 15mA MAX) , so I decided to follow the advice of removing some extra capacitors and resistors and keep it basic (like a blackface one). But nothing good has happened. I raise the value of the resistor (I went even up to 15k ohm...) on the bias circuit and the bias reading doesn't come that higher. Maximum 30mA while there was a huge red-plating developing. My tubes are ok...rather old ELECTRO HARMONIX but still ok.

Then I changed my readings to mV and when the current is 15mA the voltage is around 110mV. Pretty scary in a way.
Here's a picture of the bias circuit

https://imgur.com/a/geaiB

The pot is a 20k linear pot. Lug 1 has a 22k resistor attached , to ground. Lug 2 (center) goes to the crossroads of 2 220k Bias splitter resistors. Lug 3 goes to a 1A/1200v diode (+ side) and a 47uf/100v , the diode (- side) goes to a 1k ohm/5w resistor (the resistor I keep changing values- 1k is where i start from) and the resistor goes to the power tranformer (55v).

Any ideas where things may have gone wrong it would be much appreciated. Cheers
User avatar
xtian
Posts: 6990
Joined: Mon Apr 19, 2010 8:15 pm
Location: NorCal
Contact:

Re: Problem with biasing Twin reverb type amp

Post by xtian »

Hello and welcome! What is your level of experience?

First, while you're working on the bias circuit, pull your power tubes OUT and leave them out until we get the proper bias voltages working.

It appears from your photo that you now have a bias circuit with NO filter caps. If true, this will not work.

Let's choose a schematic to implement. Let me know if you've already got one in mind. Otherwise, let's use this:

Image
(image from Hoffman's helpful page--read the whole thing)

When your bias circuit is working properly, you should get around -45v at pins 5 of your power tubes, and it should be adjustable with your bias pot. Let's see if we can get that far before reinstalling your power tubes.
I build and repair tube amps. http://amps.monkeymatic.com
tedsorvino
Posts: 12
Joined: Tue Aug 08, 2017 9:13 pm

Re: Problem with biasing Twin reverb type amp

Post by tedsorvino »

Thank you Xtian for the quick and kind reply.
Here's another picture of the bias- rectifier board and the bias pot

https://imgur.com/a/BdqaD

My experience is pretty basic. Just a basic renovation job on pedals and new to the amps world but I ve spent quite some time on this particular project.
This is the layout I followed from their really good page.
The only thing is that my resistor is on the opposite outer lug, to the one on the layout and the value of the resistor.

My readings on pin 5 is from -33 fully counterclockwise to -66 fully clockwise. So around -45 is the average.

Thanks for the help.
User avatar
xtian
Posts: 6990
Joined: Mon Apr 19, 2010 8:15 pm
Location: NorCal
Contact:

Re: Problem with biasing Twin reverb type amp

Post by xtian »

OK, then your bias circuit sounds like it's working properly.

Redplating can happen from too much current, of course, but can also happen due to supersonic oscillation. You may want to totally disconnect your NFB circuit while you're setting the bias.

We need to know the plate voltage (pins 3) roughly, to know what target current to aim for. 350v? 400v? 450v? And you're using 6L6GC tubes? Two or four of them?

How are you reading the bias current?
I build and repair tube amps. http://amps.monkeymatic.com
tedsorvino
Posts: 12
Joined: Tue Aug 08, 2017 9:13 pm

Re: Problem with biasing Twin reverb type amp

Post by tedsorvino »

My plate voltage is aroun 430v with the tubes pluged and around 470v with the tubes unplugged.
They are simple electro harmonix 6l6 (I think they are 30w max). Usually two of them. In an amp made for 4. Placed in sockets 1 and 3. I'm aiming for 42-47mA. When I placed the 15k ohm/5 watt resistor (when the redplating happened), I went up to 22mA min. - 31mA max.With lower values of resistors it goes further down. But I'm scared to place a 25mA (I haven't got one now but I can add some smaller values in series) in order to go up around 45-50mA. Because of the redplating.
How do i disconnect the NFB (I guess negative feedback?) circuit ?
tedsorvino
Posts: 12
Joined: Tue Aug 08, 2017 9:13 pm

Re: Problem with biasing Twin reverb type amp

Post by tedsorvino »

Sorry about that. I mean to place a But I'm scared to place a 25k ohm in order to raise the bias value to the range of 40-50mA. Cheers
pdf64
Posts: 2688
Joined: Sat Mar 12, 2011 9:23 pm
Location: Staffordshire, UK

Re: Problem with biasing Twin reverb type amp

Post by pdf64 »

tedsorvino wrote: Thu Sep 21, 2017 8:55 pm... I changed my readings to mV and when the current is 15mA the voltage is around 110mV. Pretty scary in a way...
It would be helpful to explain exactly what you are measuring there, eg where you are putting the probes etc.
tedsorvino wrote: Thu Sep 21, 2017 10:12 pm...I'm aiming for 42-47mA...
Again, what exactly is this current you are referring to, where are you putting the meter probes, what is the meter set to?
tedsorvino
Posts: 12
Joined: Tue Aug 08, 2017 9:13 pm

Re: Problem with biasing Twin reverb type amp

Post by tedsorvino »

Hi pdf64. Thanks for the interest.
I 've got a "proper" bias probe- It has worked well in moste cases. So I just plug it in the socket and it goes to my meter I measure plate current mA. I'm aiming for this current, because I've already measured plate voltage and I know thetube plate dissipation. I can also measure plate voltage if needed.
tedsorvino
Posts: 12
Joined: Tue Aug 08, 2017 9:13 pm

Re: Problem with biasing Twin reverb type amp

Post by tedsorvino »

My bias probe is similar to this one: https://www.amazon.com/Eurotubes-Bias-P ... B019PAVXPG

My mistake. It juest measures current, no voltage. I measure plate voltage on pin 3.
User avatar
xtian
Posts: 6990
Joined: Mon Apr 19, 2010 8:15 pm
Location: NorCal
Contact:

Re: Problem with biasing Twin reverb type amp

Post by xtian »

So, to restate your problem, one or more of your tubes redplates if you try to approach 60-70% dissipation?

Are you sure your tubes are matched?

If the tubes are healthy and voltages are correct, the only reason I know for them to redplate is oscillation. Am I wrong?
I build and repair tube amps. http://amps.monkeymatic.com
tubeswell
Posts: 2337
Joined: Thu Mar 27, 2008 6:42 am
Location: Wellington. NZ

Re: Problem with biasing Twin reverb type amp

Post by tubeswell »

If your bias supply is working to give a steady -33 to -66V output, then methinks the problem must be coming either from loose tube socket pin clamp(s) or interaction with grid current resulting from a leaky coupling cap from the PI.

Try retensioning the tube socket pin clamps (withe th amp switched off and discharged), and then test for DC leakage on the coupling caps.
He who dies with the most tubes... wins
tedsorvino
Posts: 12
Joined: Tue Aug 08, 2017 9:13 pm

Re: Problem with biasing Twin reverb type amp

Post by tedsorvino »

Thanks again for the help guys. I'm not experienced on working an amp so your help is much appreciated.
Xtian, the problem occurs to both tubes even when I approach 40% dissipation (assuming the readings are correct and there is no problem with the probe). So I guess there is no real matching problem. How can I check for oscillation in an easy and straightforward way (without becoming too technical).

Tubeswell, there is no problem with the tube sockets. It happens to all 4 of them. How can I check for a leaky cap in the PI- in an easy and straghtforward way (I changed all caps some days ago and everything worked well)?
There are 4 caps in the area. 2 x 0.1uF output coupling , 1 x 0.1uF NFB coupling ones and 1 x 0.001uF cap - PI coupling cap.
Firestorm
Posts: 3033
Joined: Fri Jan 25, 2008 7:34 pm
Location: Connecticut

Re: Problem with biasing Twin reverb type amp

Post by Firestorm »

tedsorvino wrote: Thu Sep 21, 2017 8:55 pm Lug 3 goes to a 1A/1200v diode (+ side) and a 47uf/100v , the diode (- side) goes to a 1k ohm/5w resistor
As described, the diode is backwards. Check.
tedsorvino
Posts: 12
Joined: Tue Aug 08, 2017 9:13 pm

Re: Problem with biasing Twin reverb type amp

Post by tedsorvino »

Hi Firestorm. Thanks for the reply. I don't think there is a problem with the diode. Everything works well with the circuit. The problem must be either with the bias probe and the bias method I'm using or with a capacitor or something somewhere around the P.I. - NFB.
tedsorvino
Posts: 12
Joined: Tue Aug 08, 2017 9:13 pm

Re: Problem with biasing Twin reverb type amp

Post by tedsorvino »

Thank you guys for the great help. It was a problem with the bias probe. Absolutely unreliable. I did a test measurement across a 100Ω 1/2watt resistor on pin 8 and it seems the bias circuits is working well. I take a reading of 4.5V which means 45mA x 100. 45mA is what I'm looking for. I will change the resistor to a smaller value (1Ω /1W) in order to leave them permanently on pin 8 and measure properly.

While we are at it I ve got another question regarding capacitors. Since Tubeswell told me to check my PI - NFB capacitors for leakage and I know they are new, I decided to measure voltage around their pins.
The 0.1uF NFB Coupling cap takes about 67V and gives 0.4V and the 0.001uF P.I. Coupling cap takes about 61V and gives below 0.20mV (while the measurement doesn't stay steady). Do these measurements seem correct to you?
Post Reply