Ampeg v4 bias supply issue

[lowell]

This amp blew both fuses. I pulled the power tubes, replaced the fuses, and checked the bias. It is -1v. So I checked the bias caps and sure enough they were exploded. I replaced them. Now the bias reads -250v when the amp is on standby and -1v when fully on. I checked all the surrounding supply resistors, the bias diode and grounds and they all seem ok. Not sure what's going on here. I'm also unclear as to why the full wave bridge rectifier is grounded through a diode. ??

[lowell]

Ok I just checked the .047uf cap in series w/ the bias supply and it's shorted... put one in and voila. What's the purpose of this cap? It seems as a high pass filter w/ the 56k after it. I guess it doesn't make sense to me being that it's tapped off the PT secondary and it's already AC.

[Diablo]

If it's any consolation, I don't understand the V4 bias circuit either. Besides the .047 mfd cap mystery, try to figure where the bias voltage returns to the ground side of the transformer secondary winding when the amp is on standby. Standby disconnects the ground path, so where is that bias grounded?

There's also a mistake in that schematic for the bias circuit. The .047 cap should be connected to the junction of the 56K resistor and the diode. The above schematic shows the cap connected to ground....that won't work.

[Enzo]

Yes, the drawing added that little line from ground at the 75k resistor over to the cap. That line segment does not belong there.

There is no ground return for the bias in standby, but since the tubes are not powered, we don;t need bias then. The few seconds it takes for the bias to settle when standby is flipped won;t hurt the tubes.

[g1]

The .047 is the voltage dropping cap as they don't need the full ac from the p.t. secondary to run the bias circuit. Some models they use a resistor instead.
When yours shorted out, full ac from the p.t. got rectified by the diode and popped your bias filter caps.

[Dave Curtis, dB AudioTech]

The .047 is the voltage dropping cap as they don't need the full ac from the p.t. secondary to run the bias circuit.

Voltage does not drop across a cap (unless the cap has a high ESR). You should read the same voltage to ground on either side of it and no voltage when measuring across it. The 56K (R30) is the voltage dropping resistor. The cap in this case is a coupling cap (IMO), and if you want to make sure a shorted cap won't do this again use a cap rated for across the line or line to ground (X, Y or XY); they're not supposed to short when they fail.

[Diablo]

There is no ground return for the bias in standby, but since the tubes are not powered, we don;t need bias then. The few seconds it takes for the bias to settle when standby is flipped won;t hurt the tubes.

I understand and agree with what you are saying, but I recall that I have measured bias voltage on my V4 even when the thing is on standby....a real head scratcher.

[g1]

Sorry for the error on the coupling cap. dropping ac voltage, thanks for clearing that up Dave. Makes sense as we don't factor in wattage when dealing with capacitors.
So why is the .047 there and why did the bias filters explode when the .047 shorted?

[Gtr_tech]

The .047 cap is sort of an artificial bias tap. THe cap only passes a portion of the secondary voltage. You can read exactly how much with your meter on the side of the cap opposite the secondary winding. Mar$hall has been known to use this method to get bias on some models.

The bias filters blow up when the cap shorts.....passing more voltage than the circuit calls for. It gets rectified then on the the filters.....where they see excess voltage and heat up until they spew thier guts all over the place.

[g1]

Voltage does not drop across a cap (unless the cap has a high ESR). You should read the same voltage to ground on either side of it and no voltage when measuring across it. The 56K (R30) is the voltage dropping resistor. The cap in this case is a coupling cap (IMO), and if you want to make sure a shorted cap won't do this again use a cap rated for across the line or line to ground (X, Y or XY); they're not supposed to short when they fail.

Man am I rusty. Yes voltage drops across the cap. (this is why they use caps. in tone control circuits.) The capacitive reactance of the cap is approx. 60K (1 over 2piFC, capacitance in Farads). The cap forms a voltage divider with R30 and drops around half the voltage. You will measure an AC voltage drop across it.
As far as using an X, Y, or XY cap here, I think I would rather it short than open when it's bad. An open guarantees NO bias, a short might still give some via the diode.
Marshall used a cap. that opened when bad in some of the JCM900's. I have seen fried OT's as a result of that. (originally they used a 250V which was too low a voltage rating)

[Dave Curtis, dB AudioTech]

AC capacitor circuits : REACTANCE AND IMPEDANCE -- CAPACITIVE

and

Series resistor-capacitor circuits : REACTANCE AND IMPEDANCE -- CAPACITIVE

So, I may be a bit wrong (I was stating a generalization), but IME, I've measured the voltage across that cap in my VT22, and there wasn't any.

And caps are used in tone circuits because capacitor voltage lags capacitor current, causing phasing effects.

[imaradiostar]

Let me say it in a different way...

When you want to transfer the AC voltage on the plate of a triode to the next stage but not the DC, you use a coupling cap. This cap transfers the changes in voltage but remains charged with the DC voltage difference between ground and the plate.

When you have a full wave bridge rectified power supply the entire PT secondary is "above ground" all the time so you can't just reverse a diode to create a negative bias supply.

The bias cap "sees" the AC of the PT secondary in the same way that a plate coupling cap "sees" the audio in a preamp and passes it on to the next section.

Once of have an AC voltage across the cap but block the DC relative to the bridge rectifier, a diode can be used to charge another cap, generating bias in the traditional way that we're all familiar with.

The aforementioned formulas can be used to calculate the impedance of a cap at a given frequency into a given load. In the real world I usually just copy a given circuit and use higher voltage rated components than needed to ensure the safety and stability of the bias circuit.

jamie

[imaradiostar]

There seems to be an error in the V4 schematic- it shows the cap coming off of the secondary winding and connecting to ground and a 56k resistor. I think it should just connect to the 56k.

Am I seeing that wrong?

jamie

[Dave Curtis, dB AudioTech]

There seems to be an error in the V4 schematic- it shows the cap coming off of the secondary winding and connecting to ground and a 56k resistor. I think it should just connect to the 56k.

Am I seeing that wrong?

jamie

Correct. Enzo pointed that out earlier.

[imaradiostar]

oops, sorry I didn't read more carefully!