Marshall 100watt Plate voltage

[eddiep1966]

Wow Place in new tubes and screen voltage drop another volt Plates at 491 and screens at 489. Was checking volts drop across screen resistor to get the info to start calculating dissipation and shorted a tube socket out. What luck, but all seem to survive, but the socket and ht fuse after checking things and testing. Let me get the repair and test and Ill drop the results.

[eddiep1966]

Ok added 2.2k screen resistors
plates 496v
screens 490v
screen resistors volt drop 2.98v
g2 and k is 490v
so if i calculated right screen current 4mA and screen dissipation 1.96 watts.
monitor screen voltage hit a hard A chord screens sag to 347v the preamp is heavy modded and very hot.

[Helmholtz]

If the voltage drop measured between the ends of a 2.2k screen resistor is 2.98V, this means a screen current of 1.36mA.
Doesn't look right with EL34s. Is the meter battery fresh? Could you check with another meter?

[eddiep1966]

Not bad battery bad math. Meter is ac and calibrated a month ago.
2.98 /2,200 = 0.0013mA

490*0.0013 = 0.637mW screen dissipation.

Yell it seems off.

[sluckey]

Meter is ac and calibrated a month ago.

But you are measuring DC voltage!!!

[eddiep1966]

Ok that was my fault I did not explain that well. My meter is a Bench top Keithley it is AC powered. I'm studying lord valve pages and other sources to try to under stand this and find a solution.

[flemingmras]

Indeed, the loaded spec is 175-0-175, and so the loaded voltage should be 2x175x1.414x126/120=519.6 or in reality probably around 500V.

Not sure where the (x126 / 120) part comes from or who came up with that. But that math is quite a bit convoluted.

The AC that your meter sees when measuring AC on the transformer is the RMS voltage. Your filter capacitors charge up to the peak AC voltage, which is 1.414x the RMS AC voltage. Hence why 1.414 gets used to get the unloaded DC out.

That transformer is the standard 175-0-175 used in the full wave bridge voltage doubler circuit used in Plexi/JMP/JCM 800 100W Marshalls. Just use the 350VAC (175 x 2) -

350 x 1.414 = 494.9

Knock about 25 volts off of that for the loaded voltage and you should have about 470VDC B+...this seems to be what Marshall eventually settled on over the years.

Something else has got to be wrong. Have you measured the AC directly off the transformer on both primary and secondary? You're not plugged into a Variac with the voltage cranked all the way up are you? Most Variacs go to 140V when cranked all the way...the math seems to support this theory -

Secondary Voltage / Primary Voltage x Variac Max Voltage x 1.414 - 25

350 / 120 x 140 x 1.414 - 25 = 552.38VDC

[martin manning]

Not sure where the (x126 / 120) part comes from or who came up with that. But that math is quite a bit convoluted.

In this case line voltage was running 126 as opposed to the 120 on the data sheet

[flemingmras]

Some people call a FWB a voltage doubler, but it is two superimposed half-wave rectifiers.

It gets the FWB doubler name because of how it looks on a schematic. The CT is just providing voltage balancing across the reservoir capacitors.

As to the 126VAC...I noticed in his post he said that the wall voltage is "126 most of the time". That indicates to me that he's checked it at some earlier time, but has not checked it at this very moment.

He didn't mention the use of a Variac, but if the amp were on a Variac cranked all the way up (which would yield 140 VAC in the amp), that would certainly push the loaded voltages into the 550-560 range. With a wall voltage of 126, he'd be at about 495VDC loaded. The only other possibility is that it's a mislabeled transformer. He really needs to measure both his primary and secondary voltages and get back to us. If it doesn't conform to its published specs, he needs to exchange it. That would be the proper way to fix it.

[pdf64]

jj specify max 450v on screens, while many of the russian tubes specify 425v on the datasheets. Still, a few are 500v

If the dissipation is below the limit, eg typical fixed bias idle, then the voltage can be above the limit. With saggy supplies, things can pan out ok. The issue is both voltage and dissipation being above their limits simultaneously.
From https://frank.pocnet.net/other/Mullard/ ... 967-09.pdf

[eddiep1966]

no i was not running a variac. but measure voltage ac secondary 191.4 each. wall line 125v. Maybe I should and see how the voltage lines up.

[flemingmras]

no i was not running a variac. but measure voltage ac secondary 191.4 each. wall line 125v. Maybe I should and see how the voltage lines up.

So here is the issue -

Standardized Marshall 100W transformers typically spec 175-0-175 NLV (No Load Voltage). Hammond, for whatever reason, spec'd that as the LOADED voltage.

The No Load Voltage for this transformer is published at 185.2-0-185.2 @ 120VAC.

At 125VAC, this would calculate to 192.9-0-192.9...within less than 1% of the voltage you're seeing.

This would put your quiescent B+ at about 516VDC I'm guessing? Here is the 290HX datasheet for your reference (the No Load Voltage Chart is at the bottom right) -

290HX.png

In any case, I would definitely go with a different transformer. I have a custom wind of Metro's 1203-80 transformer (made by Heyboer) that gives you 348VAC (174-0-174) unloaded, which would put you at about 488VDC @ 125VAC (468VDC @ 120VAC, 447VDC @ 115VAC), which is a perfect variance range for EL34s. There would be a lead time as they're made to order.

[eddiep1966]

well before the tubes warm good the voltage is around 516v then as the tubes warm it falls. Now that i have it bias at 70% it settles around 494v but this is putting to much on my screens. Screens settle to around 486v. Now that i found the data sheet on the tubes max volts on screens it 425v. Any way yell time to hunt a tranny. PM me a price and info on the tranny.

[flemingmras]

well before the tubes warm good the voltage is around 516v then as the tubes warm it falls. Now that i have it bias at 70% it settles around 494v but this is putting to much on my screens. Screens settle to around 486v. Now that i found the data sheet on the tubes max volts on screens it 425v. Any way yell time to hunt a tranny. PM me a price and info on the tranny.

I gotta call in another order tomorrow so I'll get a price on it then. I'll DM you my info.

What is the DC voltage at the HT fuse with all tubes out and the standby switch set to On/Play (asking for curiosity purposes)?

I have seen some homebuilt Super Leads with these voltages. I built one back in 2004. The first generation of MetroAmp's 69 Super Lead Plexi kits used a Heyboer HTS-6010 power transformer with a 188-0-188 secondary that got about the same voltages you're seeing on your Hammond. But many people experienced red plating on the power tubes with them. On the older transformers, the regulation wasn't as tight as the ones made today. They would sag enough to not be an issue. When I spec'd my custom wound one, I spec'd it off of a JCM800 2210 that just happened to be a couple of volts lower than most JCM800s I've seen. The amp sounded amazing, so I had Heyboer clone that transformer's B+ winding spec.

[Roe]

well before the tubes warm good the voltage is around 516v then as the tubes warm it falls. Now that i have it bias at 70% it settles around 494v but this is putting to much on my screens. Screens settle to around 486v. Now that i found the data sheet on the tubes max volts on screens it 425v. Any way yell time to hunt a tranny. PM me a price and info on the tranny.

As PDF64 explains above "If the [screen] dissipation is below the limit, eg typical fixed bias idle, then the voltage can be above the limit. With saggy supplies, things can pan out ok. The issue is both voltage and dissipation being above their limits simultaneously."

So the 425v limit does not matter too much. But you need 1k+ screen resistors