Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: texwest on December 14, 2019, 01:16:18 pm
-
Hey guys wanted to get your advice on this. I'm working on this '70 superbass marshall with 518 on the plates and 514 on the screens. I installed new svetlana el34's and within minutes had a tube failure that took out the tube and the 1k screen resistor. I had it biased at 30 ma from the beginning so no problem there. The other tubes are ok for now.
The svetlana is rated at 800v on the plates and 500v on the screens.
Should I be considering lowering the screen voltage?
Is there a new production tube that can take this?
Thanks ahead of time!
-
You installed four new tubes? And only one failed? Maybe it was just a rotten egg. I'd probably try another svetlana. Hopefully some of the guys with real EL34 experience will chime in.
-
Like sluckey said, it could have been a bad tube. Infant mortality is not a rare occurrence for new production tubes. Where did you get them?
Also, Marshall increased the screen grid resistor value to 1.5 K in the JCM 800 series amps. (Misread the schematic, old eyes, cheap readers) This still might be a good mod to consider so as to prevent the screens from drawing excess current.
-
I read somewhere today that marshall once had 2.2 k on the screens during the 90's. So 1.5k still might be a good idea.
-
Those voltages/differential (plate/screen) are basically what is listed in the Marshall archives - not too far apart. Reminds me of my Major that absolutely tortures tubes. Something else to consider are those resistors are almost 50 years old. They certainly could have drifted and the resistor may have allowed the event to occur before becoming collateral damage. Check for discoloration on the others and maybe check values. Certainly bumping up to 1.5 or 2.2 would allow for more civilized operation. Tubes are not what they used to be.
Jim
-
I read somewhere today that marshall once had 2.2 k on the screens during the 90's. So 1.5k still might be a good idea.
JCM 900 series.
https://el34world.com/charts/Schematics/files/Marshall/Marshall_jcm900_21xx_25xx_100w.pdf
I seem to remember reading a Mullard publication at some point in time that listed a range of screen grid resistor values that was based upon screen grid voltage, and the 1.5K value was listed for higher voltages. But then again my memory may be failing me.
-
Thanks for your replies!
I'm thinking that the real solution is to always run this amp on a variac.
I just did this at 117 volts and the heaters were exactly 6.3, plates to 493, screens to 486.
What do you guys think of those voltages?
-
A 120:6V 2A stepdown would give you "right" voltages.
However I can't believe a 5% drop makes a difference between "within minutes" and perpetual happiness.
I also think you just got a dud.
Larger screen resistors are always a good idea.
-
You also have a 110V tap and a 120V tap. Make sure you are on the 120. It may have come to the US on 110 back then or some clown thinking higher voltages will make him sound just like Eddie.
Jim
-
Thanks I have now checked the taps. It took me awhile to find the color code chart.
-
Hi guys (had some experience with the JCM800 and 900)
Around 20 years ago, we went through a real drought (in Australia) when it came to tubes especially EL34s. I understand most tubes including "handpicked MESA" were Chinese and soon a lot of my friends started getting NOS tubes from OS.
The common local brand was "RUBY" and very soon I discovered that for every quartet used on Marshalls, I needed a spare twin set. Common problems, redplating, the odd arcing inside and generally unreliable and did not last. Fortunately 6L6's were usually Svetlana or Sovtek and quite reliable.
I have recorded voltages as high as 525V on Plates, Bias around -42V on a Model 2203 (JCM 800) fitted with EL34 quartet, with mains AC (Canberra, Australia) at 247V AC (50Hz). I have rarely replaced a JCM800 output Tx and Heater voltages were very close unlike later model Marshalls.
I found No real faults with these amps (makes you nervous :icon_biggrin:) but it was a battle to get past the first few hours (of soak testing) to see if the amps were going to last.
So I'm with PRR and Sluckey, you may have got a dud EL34, assuming Screen Grid Volts and resistors are OK and Bias Volts/Grid resistors stable. Just as a matter of interest I see a lot of Fender Deluxes, Vibrolux, Champs, etc and most run HT at 10% to 20% above schematic values and the amps are surviving OK. (I do pay attention to bias and set for correct Plate dissipation :laugh:)
The supply of EL34 is much better these days, so I see fewer problems.
Trust this helps.
Kind Regards
Mirek