Peavey Classic 400 schematic ?

[The Dude]

Gotcha! I missed that part (obviously).

[nsubulysses]

Good news an SVT can do 400W also. They just don't call it that so you don't get confused


Is this all because of the 400R dropping resistor in series with the screen supply in PV classic 400??

[J M Fahey]

[ATTACH=CONFIG]45784[/ATTACH]

Maybe I am at war with reality

I have an SVT-II here so I might as well give it a look for comparison

here's how it does 34V into 4 ohms

B+ is 655V and drops to 565V at clipping. Screen supply drops less compared to Peavey Classic 400, 385V at idle and 327V.

SVT power consumption is 220W at idle and 650W doing 34V into 4 ohms.

SVT power tubes (I only measured one) are doing 132mA when amp is clipping

Maybe this Peavey classic 400 is different because it has a saggier screen supply and 2 additional power tubes so it does similar power to SVT but is supposed to be easier on the tubes.

Seems disapppointing but possibly real??

No, that´s not war but agreeing with reality.

There´s 3 points to be considered here:

1) it is "assumed" that "if 6 tubes give me 300W, then 8 must give me 400" ... clearly not the case.
Tubes are *one* factor in the equation, but there´s more, with the same level of importance:

2) as certain "Enzo" once said: "the amplifier is that thingie between the power supply and the speaker"
IF Peavey PSU drops more than Ampeg then so be it.

I always say that tubes are strongly current limited, so in principle Peavey "should" have 33% more peak current available.
Fine ... if supply can supply it, if screens make tube pass it, and grids can be fully driven.
If not ....... we only have 2 extra tubes, which look killer but maybe are not fully used.

3) having the tubes is fine, but fully driving them is just as important.

Waveforms show that Peavey is struggling to drive those power grids, there´s lots of poorly named "crossover distortion" which actually is grid rectification and bias shifting.

Remember that RMS voltage is one "averaging" parameter, considering the area under the curve.

Compare both waveforms: even IF they were to reach same peak voltage into the load, area under the Peavey waveform is peakier, narrower and smaller than area under the fatter rounder Ampeg waveform.

Peavey area being smaller (means less RMS even if everything else is same) than Ampeg one.
Or in this case, having 2 extra tubes sort of compensates poor waveform but provides about same RMS power, not 33% more as expected.

Ampeg waveform can brag of effortless drive, not very common in tube power amps which require exaggerated overbiasing to compensate for that.

[J M Fahey]

This is what I mean by "more area under the curve":
I poorly drew 2 half "sinewaves", the red one being what it becomes when amplified by most tube amps, unless Class A or at least heavily biased at idle.
Look at the red painted area.

The other, more perfect sinewave, is what all SS amps and a few tube ones (such as Ampeg SVT) put out.
For the exact same width and peak current/voltage , area under curve, which is Red PLUS Green, is clearly larger.
There are "more electrons during more time" pushing those voice coils and obviously, both Average and RMS power are higher.
Please don't nitpick that "RMS power does not exist" and such, we are using the popular name here.

By the same token, that´s why squarewaves provide highest power (and also burn speakers), "area under the curve for a square wave is maximum possible": "all the current possible during all the time, continuously"

Squarewave area under the curve is red + green + blue: maximum possible.

[g1]

Maybe this Peavey classic 400 is different because it has a saggier screen supply and 2 additional power tubes so it does similar power to SVT but is supposed to be easier on the tubes.

Both have separate windings for screen supply. PV has 400ohm common screen resistor that SVT does not. And only 50uF filtering vs 100uF in the SVT. So these may account for some difference.
JM showed how the PV waveform is peakier, leading to lower RMS readings. The only question is whether the peakier waveform is normal or not.
I'm not opposed to these only being capable of 300W, except PV is usually pretty honest, and, as Jazz P Bass mentioned, they say 420W at clipping.
The only other things you might want to check:
SVT is biased around 17W per tube. Not sure about the PV, you mentioned -53V but the only thing I see listed is -50V ?
You said the 12AT7 phase inverter had shorted. The AT7 is actually the driver, and it does need to drive 4 power tubes per triode. You checked all the associated resistors that could have been stressed by the shorted AT7? Have you tried any other AT7's?

[vintagekiki]

B+ is 655V and drops to 565V at clipping ...

If B+ (after Rectifier) at clipping drops from 655V to 565V how many drops V~ (before Rectifier).

If the V~ at clipping does not significantly drops (1-2%), it is highly probable that the elco capacitors C100-C106 has changed the value due to age or has been partly "dry".

If the V~ drops, the problem is in the wire thickness of the primary or secondary for the B+ power transformer.

Expected output power depending from B+ can be approximately calculated P = (B+)˛ / Raa.
In the above example, if B+=565V and Raa=1kΩ output power is 565V˛ / 1kΩ = 320W

[nsubulysses]

If B+ (after Rectifier) at clipping drops from 655V to 565V how many drops V~ (before Rectifier).

If the V~ at clipping does not significantly drops (1-2%), it is highly probable that the elco capacitors C100-C106 has changed the value due to age or has been partly "dry".

If the V~ drops, the problem is in the wire thickness of the primary or secondary for the B+ power transformer.

Expected output power depending from B+ can be approximately calculated P = (B+)˛ / Raa.
In the above example, if B+=565V and Raa=1kΩ output power is 565V˛ / 1kΩ = 320W

It's interesting you say this because my customer was told by another shop that "The issue {repair shop name) said was causing the fuses to blow was that the power transformer wasn’t able to keep up with what the tubes needed. Not sure if that is relevant."

I did not think much of it because it seems like if the amp was current limited by the transformer it would be HARDER for it to blow fuses. But anyway I am still at war with reality and will look at this more with G1s and Vintagekikis suggestions, and I also emailed PV

Juan, you are a mastermind but have you ever personally scoped one of these amps? Someone must have. PV is gonna settle the score on if this amp can do 400W at clipping. I sent them the scope images so hopefully they will say be able to say if thats normal or not

G1, i did try diffferent 12AT7s and also tried to find any burnt or out of spec resistors in those gain stages but no luck. Maybe I should re-check just in case.

[g1]

Can you get an idle current measurement for the power tubes? Maybe the method where you measure resistance and voltage at OT primary halves?

[nsubulysses]

OT primary resistance per side is 10 ohms

At idle I get .45V and .42V per side so about 10.5mA idle current per tube

At 34V into 4 ohms the sides do uneven power. One does 23V and the other side does 30V, so 57.5mA and 75mA per tube per side roughly. so it seems like at tube or two is weak and not there is not a lot of current draw of the phase inverter can't drive them. So putting in different power tubes and 12ax7 and 12at7 it still does the same. maybe I missed something in the phase inverter.

Unfortunately did not get to do the vintage kiki test yet and now I gotta go won't be able to check this agian til saturday

[J M Fahey]

I have not even *seen* a PV400 in Argentina, people here goes straight to SVT, which are abbundant.
Guys buy what they see on famous bands backline.

[vintagekiki]

Biasing Power Tubes

http://web.archive.org/web/20100923051727/http://www.diyguitarist.com/GuitarAmps/PT-Biasing.htm
Biasing Power Tubes

http://web.archive.org/web/20120507060924/http://www.diyguitarist.com/GuitarAmps/PowerTubeBias.htm
Power Tube Bias Charts

http://web.archive.org/web/20110621211356/http://diyguitarist.com/Images/BiasChart-KT88.jpg
KT88/6550 Bias Chart

http://www.duncanamps.com/technical/lvbias.html
How to bias your amplifier

http://web.archive.org/web/20080521090437/http://www.webervst.com/tubes/calcbias.htm
Bias calculator

... At idle I get .45V and .42V per side so about 10.5mA idle current per tube ...

Idle current is measured on the cathode. Since tubes are matched, the idle current set for one tube.
If the idle current 10.5 mA measured on the cathode of tube, I personally think that 10.5 mA is a small idle current, it is necessary to set with the VR100 to 20-25 mA per tube

[nsubulysses]

the smoking gun.....

I emailed PV and described my dilemma of basically "this thing seems to not work quite right but I can't' find anything wrong with it" and I sent scope images and a lenghty explanation.

PV has responded:

Sorry for the delay as I had to get help from engineering. This is what I'm being told.....


The output transformer was designed around these weird 6L6-based 6550s, which were the only decent 6550s available at the time. What the design daddy of this monster did not know at the time was that the plate impedance was a little higher than the accepted 6550 spec. Consequently, the Classic 400 will only put out 400W into a mismatched load with normal 6550. The 8 Ohm tap "wants to see" 12 Ohms, 4 Ohm tap wants 6 Ohms, etc.. you can mismatch the load on the bench to see this in action. Full power can be achieved in actual use at the expense of damping factor, which sounds AMAZING when you put 8 Ohms on the 4 Ohm tap.

The only other option is to find a set of those ca. 1996 tall-based Chinese 6550s or check the plate impedance on some other KT88 variants and try those.


So basically what you are seeing is normal. The initial design guy is no longer with us who by the way, wanted to name the amp "Crankenstein 400" which would have been pretty cool.
I hope this helps,
G. B.

[vintagekiki]

DANGEROS HIGH VOLTAGE
AMPLIFIER MUST BE DISCONNECTED FROM AC MAIN 117/230 V

How much is Raa when the load is 8 or 4 ohms ?

Check the plate impedance
On a secondary 8 Ohm connect a known voltage (e.g 10V ~) and measure the total voltage between the output tubes anodes.

By ratio Uaa~ and 10V~ voltages and simple account gets the approximate value of Raa.
Repeat the same procedure with secondary 4 Ohm
The difference between the values of Raa for 4 and 8 ohms should be a few %

Plate impedance for 6550 is in the order of 4000 - 5000 Ohm (for 2 tubes PP) or 1000 - 1250 Ohms (for 8 tubes PP).

Based on empirical choice plate impedance (eg 1000 to 1250 Ohm) and measured ratio (Uaa~ and 10V~) calculation Rzv.

>> Voltages between the anodes of power tubes and +HV (middle point primary OT) must be symmetric.
>> Asymmetricity greater than a few % indicates a partial short circuit between the primary windings.

[J M Fahey]

Mmmmhhhh, shorted turns won´t affect just the winding they are on, but all of them at the same time, so you won´t "create" an assymetry if it wasn´t there before.

The transformer core couples all windings to each other.

If you have shorted turns, the whole transformer will be lossy, no matter what winding you measure.

That said, it looks like published spec will not be met with current 6550, unless some odd load impedance is applied.

Fine with me, that amp is a BEAST any way you look at it and in any case it can put out *somewhat* higher power than a comparable SVT.
I woud not lose sleep over it.

I have found many perfectly adequate 250W RMS or so SVTs, and the odd 200W one, where retubing is advised but still can be used as is, for a little more time at least.

[vintagekiki]

shorted turns = partly short-circuit primary winding.

This happens if the amplifier is left without a speaker. the spark jumps between the winding.
When the ohm meter is measured it will be difficult to notice the difference in resistance, but when the transformer is reversible, the voltage difference +Hv/Ua1 and +Hv/Ua2 can be higher than 5%.
This is noted in the impossibility that the amplifier reaches its rated power.

If the OT is correct, the most likely power tubes are not matched for the same bias, so they are unequally opened (unequal clipping of the upper part of the sinusoid)