redplating 1959 w larmar

[novosibir]

how do I add the 100k in series with v3a's grid?

Simply think, that you use the 100K instead of the green cable to V3's grid pin aka at the end of the cable. It's a so called swamp resistor, what should be as close as possible at the tube socket.

Larry

EDIT: RA was 3 minutes quicker than me

[Roe]

with a 100k tail resistor, the cold bias didn't give me any redplating at all. with hotter bias (65% diss) V6 (not v5) started to get a very, very slight redish colour after a while.

EDIT: the stock PI appears to be blame since it drives v4-5 much harder than v6-7

[Roe]

with a 100k grid resistor on v3a (I added a 56k in series with the 47k that was already there), nothing changed:
with a 10k tail resistor, and the 65% bias, one RTF (V5) still redplates.

[novosibir]

The downside of a large tail resistor although is, that you're diminishing the headroom of V3 and the gain.
To avoid or balance this, you could raise up the supply voltage to V3's node by bypassing one of the two 10K's in the HV rail, but then you'd have to increase the drop resistor's value from V3's node to V2's node accordingly, to not have a too high supply voltage to V2 & V1

If you raise up V3's supply voltage, keep an eye on the voltage rating of the output couplers, to not exceed their limits! The lowest value must be at least V3's supply voltage plus the value of the bias voltage, both at idle.

An example: The output couplers are connected between -45V (bias) and +370V (V3 supply) - in this case the difference is 415V, what means, that a 400V cap might (and will anywhen) fail

Larry

[raiken]

with a 100k tail resistor, the cold bias didn't give me any redplating at all. with hotter bias (65% diss) V6 (not v5) started to get a very, very slight redish colour after a while

That says you indeed got a symmetry improvement. However, that also limits the headroom of the PI, so it will clip sooner, although it should still clip later than the output tubes.

This can be a good thing, because it will keep your crossover distortion from getting out of hand on heavy overdrive (it limits the "buzz" that rides on the note decay).

RA

[Roe]

I guess the readings across the HT fuse look ok?
it measures .15A at idle and .5A or even .6A when crancked

[raiken]

Aiken, this brings up some interesting thoughts here. Some math I was playing around with earlier...not sure if it's correct and if it isn't please feel free to correct it -

Assuming a 100% efficient zero loss output transformer and single frequency test signal conditions into a resistive load -

With each 1/2 of the OT primary being 400 ohms (roughly), you would need 500mA draw through that 400 ohm load, which would in turn create a 200 volt drop across the 400 ohm load, to get 100 watts on 1/2 the OT primary, which then would get transferred to the secondary.

However, no transformer is even close to 100% efficient, which means that you'd have to be feeding MORE THAN 100 watts of power in to get 100 watts out, which would require more current and therefore a higher voltage drop across each 1/2 of the OT primary.

And of course, as current flow through the tube & the load increases, so does the voltage drop across the load, which causes the voltage drop across the tube (from cathode to plate) to decrease as current flow increases. Effectively, the PI grid drive is modulating the cathode - plate resistance of the tube at input signal frequency. As plate - cathode resistance decreases, plate/load current increases and hence the decrease in plate voltage, which is what puts your plate voltage/current out of phase. Effectively you have the exact same thing happening in the bias supply right at the bias pot as you adjust it.

Am I correct in thinking this?

First, I think your math is a bit wrong because you didn't take into account RMS for power, and you're using the wrong impedance. The way to look at power is this:

The plate signal must swing along the loadline, so the plate voltage swing will run from roughtly 480V at cutoff, down to around 50V at the point where it clips at the zero volt grid curve. This gives a voltage swing of 430V peak (not peak-to-peak, because we are only looking at one half of the sine wave in a class B, or cold-biased class AB amp).

Next, you look at the current - we have a current swing of roughly 500mA for that voltage swing.

Now, these voltages and currents are peak, so we have to convert them to RMS, which means dividing by a factor of sqrt(2), or 1.414.

You could multiply the RMS voltage by the RMS current to get power (the square roots cancel, so you just divide by two): P = (430V*500mA)/2 = 107W


Since we can determine power using P=I^2*R, the power for one tube pair will be (500mA/1.414)^2*850 = 106W.

Of course, this is assuming an ideal tube, ideal transformer, etc., so you won't get that in real life.

RA

[raiken]

It just so happens I have a JVM 100W amp in here that a friend asked me to repair. It has 500V on standby, 480V on the plates at idle, and sags to...get this...408V at full clip! The screens sag from 466V at idle to 340V at full clip! Of course, the screens use a 470 ohm dropping resistor instead of a choke, so that is contributing a bit.

Compare this to your 30V sag at full clip...

I am measuring right at 100V p-p into 16 ohms, for an output power of around 80W when the clean channel hits clipping.

I guess that's one way around the redplating issue...

They don't make 'em like they used to...

Edit: BTW, in troubleshooting the JVM, I found a blown HT fuse (internal, 1A 5mm pain-in-the-butt), an open 1 ohm cathode resistor with no sign of burning - the guy was running on one side of the push-pull pair only! After I fixed that, I also found a pin on the PI tube that had no solder on it at all - it was making an intermittent connection.

It was odd, because it would work with a JJ tube or the original Marshall tube, but not with an EH, or Ruby, and it would occasionally cut in and out on one side. With the EH or Ruby, the output was cut off on the top, like a half-clipped waveform, because only one side of the PI was working when it wasn't making contact.

The guy was complaining that the entire amp would cut out from time to time. I think the intermittent PI was cutting in and out on the one side he was running!

Randall Aiken

[Mars Hall]

Does this mismatch apply to 50 watters as well? I plan on purchasing a bass spec 50 watt Metro kit in the future. Will I need to swap the PT for one with lower voltage?

How much voltage do you get out of the PT? I don't think any 50 watters run near 480V...my '72 ali panel head sitting behind me only runs 410V on the plates (and sags to about 380V at heavy overdrive), with a 3.4K primary. I think the highest 50W voltage I recall seeing was a '69 plexi that had 445V on the plates. I didn't measure the plate-to-plate impedance on that one, unfortunately, but it was probably 3.4K as well.

Randall Aiken

According to the wiki, the Metro 1202-118 PT is putting out around 440v B+, will this be too hard on current production EL34's and cause the same redplating issue seen on the 100's? I'm only concerned about how this will work in a Metro kit.

http://metroamp.com/wiki/index.php/1202-118" onclick="window.open(this.href);return false;

[raiken]

According to the wiki, the Metro 1202-118 PT is putting out around 440v B+, will this be too hard on current production EL34's and cause the same redplating issue seen on the 100's? I'm only concerned about how this will work in a Metro kit.

http://metroamp.com/wiki/index.php/1202-118" onclick="window.open(this.href);return false;

\

No, that should be okay, especially if the 440v rating is at idle, because it will sag at full power. The problem with the 100W is that it apparently only sags to 450V, which, coupled with the asymmetrical drive from the phase inverter, can exceed the dissipation limits of the tubes.

RA

[Mars Hall]

No, that should be okay, especially if the 440v rating is at idle, because it will sag at full power. The problem with the 100W is that it apparently only sags to 450V, which, coupled with the asymmetrical drive from the phase inverter, can exceed the dissipation limits of the tubes.

RA

Ahhh, Thanks! Now I feel all warm a fuzzy inside again.

[flemingmras]

It just so happens I have a JVM 100W amp in here that a friend asked me to repair. It has 500V on standby, 480V on the plates at idle, and sags to...get this...408V at full clip! The screens sag from 466V at idle to 340V at full clip!

Compare this to your 30V sag at full clip...

I am measuring right at 100V p-p into 16 ohms, for an output power of around 80W when the clean channel hits clipping.

I guess that's one way around the redplating issue...

They don't make 'em like they used to...

Randall Aiken

So in other words the regulation of the supply in a JVM is worse than in a vintage Marshall!? What did they do, make the supply a lot less stiff? Or is the current output rating of the transformer underrated for the amp?

BTW thanks for the correction on my math.

[novosibir]

So in other words the regulation of the supply in a JVM is worse than in a vintage Marshall!?

No, not necessarily!

I have a '68 Plexi Super Bass (serial #12284), absolutely in orig. condition with orig. irons in it, aso. - except I've made a cap job a while ago, because the orig. caps were junk.

With the voltage selector set to 240/250V and a wall voltage of 230V (Germany) I get a B+ of 480V at idle. Full blast the B+ is sagging down to 390V - the screens supply I haven't measured.

Larry

[raiken]

So in other words the regulation of the supply in a JVM is worse than in a vintage Marshall!?

No, not necessarily!

I have a '68 Plexi Super Bass (serial #12284), absolutely in orig. condition with orig. irons in it, aso. - except I've made a cap job a while ago, because the orig. caps were junk.

With the voltage selector set to 240/250V and a wall voltage of 230V (Germany) I get a B+ of 480V at idle. Full blast the B+ is sagging down to 390V - the screens supply I haven't measured.

Larry

I think the answer is that there is no answer - they are all over the map. This is why some 100W amps redplate, others don't.

This Metro iron must be really stout to only sag 30V!

RA

[Roe]

could caps affect this as well? I'm using the sozo 4 100uf caps for screens and mains (for a total of 50uf on mains and 50uf on screens). PI caps is a F&T 32+32uf and preamp F&T 16+16. all caps have been slow-formed according to larry's method (a 100k after rectifier)