Ampeg V4 Bias Issues

[AHeck]

I'm working on my Ampeg V4 and I'm getting a considerably higher Bias voltage of -62VDC from 253VDC instead of the specified -52VDC.  I've checked resistor values and they are very much inside of tolerance. I've included the circuit schematic and would not only appreciate guidance to get the voltage in spec, but also a basic explanation of how the circuit works to drop the voltage.
As it stands now, I am running two 6550s with the out impedance switch on the amp set to 4 OHMS and my Dummy Load Box set to 8 OHMS to compensate for only using the two output tubes and I've changed the screen resistors to 1K.  The math I've done indicates that each tube is cruising at 100mA, or 120% dissipation.
I wouldn't think that the types or number of tubes being used would have any bearing on the bias voltage, but I'm often times wrong, so please school me on how I'm wrong.  I've a matched quad of 6L6GCs in transit and would feel foolish if I were to make any changes that were unnecessary considering the imminent tube swap.
The reason for only using the two tubes is that the amp (which works perfectly in a rehearsal setting) has twice blown the 10A internal fuse during gigs taking out one tube upon each occurrence, two gigs at the same venue actually.
Any correspondence would be welcome.  When functioning, the amp is very, very dead quiet and has an amazing tone, but that's all for naught when not functional. 
Thank you all, very much.

[AHeck]

I'm not able to post the schematic right now, but will try later.  Sorry

[shooter]

I'll double check, but -62 is a colder bias than -52
the closer to zero you get the hotter the bias.
what parameters did you use to do the math?

[AHeck]

what parameters did you use to do the math?

The resistances from the used pins 3 to output transformer center tap both came in at 61 OHMS and the Voltage drops associated both came to 6.1VDC.  Did I miss something with only running one tube per side? Thanks for the reply.

[shooter]

here's the schematic link
Ampegv4poweramp.pdf (el34world.com)


with the PA tubes out, AND the amp in standby
measure the vdc from pin 5, tube side of socket, to chassis ground


IF it's my amp, I would put 1 ohm resistors from PA tubes, pin 8 to ground to make measuring tube current a safe, no-brainer.

[AHeck]

with the PA tubes out, AND the amp in standby
measure the vdc from pin 5, tube side of socket, to chassis ground

487V

[shooter]

 
guessing your 4 and the one I linked ain't the same.
pin 5 on the schematic I linked is G1, where the bias volts wind up.
how bout re-doing with whatever your PA tube grid is

[sluckey]

A better guess might be... WRONG PIN!   

[AHeck]

A better guess might be... WRONG PIN!   

No sir,  I'm probing pin 5, when I flip the standby switch closed, voltage at pin 5 goes from 487vdc to -62vdc.

[AHeck]

Looking at the schematic, and please forgive/correct me if I/m wrong, with the standby switch open, the 56K and 75K resistors and the two 10uF caps from the bias circuit disappear leaving only the .047uF coupling cap, the rectifier diode and a 10K resistor in series with the 253VAC from the PT and the 100K resisters in advance of the grid. 

[sluckey]

A better guess might be... WRONG PIN!   

No sir,  I'm probing pin 5, when I flip the standby switch closed, voltage at pin 5 goes from 487vdc to -62vdc.

Oops. Sorry.

[shooter]

Looking at the schematic,

I missed the float when the amp is in standby.  with the switch open, there is no ground for bias supply, so the tubes "float" to B+
I think, it's early and I'm hungry!

[pdf64]

Schematic link?

[shooter]

reply #4

[pdf64]

Ahh, I see the issue now, I’d assumed that the standby switch was in the positive supply feed.

[shooter]

with that configuration, if the caps are from the 70's,'80's, maybe early '90s, I can see how it could pop a fuse.

[AHeck]

Thanks for sticking with the thread. I've replaced all the power supply caps and had very good performance from the amp for several months.

[AHeck]

Also, I'm pretty jazzed that I was apparently kinda right about the bias supply, but isn't putting all of B+ to the grid weird?

[shooter]

B+ isn't on the grid.  The grid has no "reference" since ground is "turned off"
you will find B+ on everything until ground gets turned back on

[66Strat]

B+ isn't on the grid.  The grid has no "reference" since ground is "turned off"
you will find B+ on everything until ground gets turned back on

Where is the ground reference? The only ground references that I see are the power tube cathodes and the  output transformer. My old eyes see the bias referenced to the neutral leg (roll the dice) of the power transformer AC winding.

[sluckey]

Where is the ground reference? The only ground references that I see are the power tube cathodes and the  output transformer. My old eyes see the bias referenced to the neutral leg (roll the dice) of the power transformer AC winding.

See pic. The FWB negative side is the power ground. Plates and screens are always hot. The other green grounds are controlled by the STBY switch.

[66Strat]

I don't understand the ground scheme. Is this a floating ground? On my Fender Concert, the FWB rectifier has a chassis ground reference as indicated in the circle. I don't see a chassis ground reference on the SVT. The ground reference controlled by the standby switch goes to the ground switch with a .047 uF capacitor to block DC.

[DummyLoad]

I bypassed the standby switch on my V4 for same reason. IMO, disconnecting OT CT would have been best for these amps.


--Pete

[pdf64]

I don't understand the ground scheme. Is this a floating ground?…

That schematic shows a 2 prong era amp.
The one below shows a 3 prong era amp, with a mains safety earth, and also the chassis connection that’s not marked on the previous schematic.
https://el34world.com/charts/Schematics/files/Ampeg/Ampeg_v4_vt22.pdf

[shooter]

maybe there were some flameouts between the '71 version and the '77 one  , or Nader was in the building

[pdf64]

So if there’s 487V on the grids in standby, what’s happening to the bias supply caps?

[shooter]

the 487v is "fantom" voltage i believe, since the Bias supply has no ground until the switch closes.

[pdf64]

I think it’ll depend where exactly the meter is connected to, ie the switched HT supply common or chassis common ( = the bias supply common).
To go up high, I think it will have been connected to the switched HT common.

[shooter]

it hurts my brain to think that hard 


all I know from experience, when "everything" measures B+, ground is missing somewhere.

[sluckey]

all I know from experience, when "everything" measures B+, ground is missing somewhere.

Of course it's missing. The STBY switch did that.   

[66Strat]

In the schematic posted by Shooter in reply number 4 there is no chassis ground reference. This schematic was for a 1971 V4 drawn by Joe Piazza. Was this an error in the schematic?

The schematic posted by pdf4 in reply number 23 for a 1977 V4 shows the chassis ground connection.

[pdf64]

In the schematic posted by Shooter in reply number 4 there is no chassis ground reference. This schematic was for a 1971 V4 drawn by Joe Piazza. Was this an error in the schematic?
…

I think it must have been.
The node that sluckey used the green highligher on in the attachment in reply #21 representing chassis common.

[66Strat]

I don't see it. The schematic that Sluckey highlighted in post number 20 is the same one that Shooter posted, hence my question regarding floating ground in post number 21.

[shooter]

I got the stby switch, what i'm strugglin with is explaining "why" G1 measured 487.
I'm not a real engineer, i just played one in the field 

[sluckey]

I wonder what that voltage would read if there was a 22K resistor tied across the meter probes.

[shooter]

 
This thread inspired this thought.  un-hijacking

[AHeck]

This has been very interesting and I'm wondering if there are any tests I could do here to help answer the question of why there are 487V on the grids such as testing using different ground points with standby and the ground switch in different states.  I'd also still like to know why the two 6550s are running at 100mA, or even if that is correct. Regardless, thanks for showing up and helping out.

[PRR]

> testing using different ground points

The only thing the tube cares about is the Grid-to-Cathode voltage. In the V4 Standby the cathode is also near 487V and the tube feels "no" voltage. (Some measurement error due to meter loading.)

As for the 100mA... put 1 Ohm resistors in the cathode circuits, then you'll Know.

[pdf64]

I think this is the original Ampeg V4 schematic

With reference to chassis common, I can’t see how the pin 5 grids could be up at high V DC in standby 

Could that be rechecked?

To cool down idle anode and cathode current, the value of R50 or R49 could be increased.

[shooter]

there is NO high VDC on pin 5 in standby, the meter however with it's huge 10meg impedance(?) thinks there is.
Sluckey mentioned a 22k between black and red while making the measurement.  doing that should make the meter believe in reality.

[pdf64]

My concern is due to reply #6, which indicates 487V from pin 5 to the chassis.
The schematic shows a total of 222k of resistance between pin 5 and the chassis, with the cathode being directly connected to chassis.
I agree that the grid can’t be 487V with respect to the cathode; if it was stray voltage with no current sourcing capability though, the grid would be forward biased and shunt it to the chassis. Hence to me, things don’t seem to square up, so it would be nice to confirm that pin 5 standby voltage, with respect to the cathode / chassis common.

I suspect that the reading was taken with respect to the HT rectifier negative terminal node (ie the switched common of the HT supply).

Whatever, it’s a very good idea for the control grids to have their normal bias voltage whilst in standby. As otherwise, the valves will momentarily pass heavy current when standby is flipped to operate mode. Due to it taking a few seconds for normal bias voltage to be established.

[shooter]

I understand your concern, i've chased many rabbits in my life!  I suspect even the design engineers figured it out by '77 based on the schematic you posted.
It really boils down to do i want to re-design the amp, or keep it '71 original.  that is what folk like us get paid for "Hey can you....sure $19.99"