Marshall JMP 2203 - 6550 Bias

[Cap217]

Ouch, your plate voltage should not be that high at 116V to 117VAC coming in. Sorry missed where you mentioned this at the original post. something is a bit of a amiss here, IMO

Thats fine (ps: My response wasnt a disrespectful one, it may have come off that way in reading). But if I would let the 123v into the amp Im sure we would be at 530 or more.

I do have another amp, a 72 super lead that has 500 plate voltage as well. Its interesting how they vary.

 

[Racerxrated]

I’ve always read that to be safe, since Marshall transformers are designed for EL34 NOT 6550s, you should bias them as an EL34….so between 30-40ma. If you bias them in the 50+ range like a typical 6550 you’re taking chances on stressing the transformer. The higher plate voltage PTs appeared in 1982, after dropping down in 76/77 in my experience(early 70s Marshalls with the 2562 Dagnall could be 520! My 72 has a 3556 which is 485pv). I’ve experimented with GE 6550s in JMP/JCMs and I didn’t hear a better tone between 35ma or 55ma. So 35 is where they stayed in the last 6550 Marshall I owned. Longer tube life and transformer safety for the win.

 

[Cap217]

I’ve always read that to be safe, since Marshall transformers are designed for EL34 NOT 6550s, you should bias them as an EL34….so between 30-40ma. If you bias them in the 50+ range like a typical 6550 you’re taking chances on stressing the transformer. The higher plate voltage PTs appeared in 1982, after dropping down in 76/77 in my experience(early 70s Marshalls with the 2562 Dagnall could be 520! My 72 has a 3556 which is 485pv). I’ve experimented with GE 6550s in JMP/JCMs and I didn’t hear a better tone between 35ma or 55ma. So 35 is where they stayed in the last 6550 Marshall I owned. Longer tube life and transformer safety for the win.

I am at 55% of a 6550 so thats 23 watts. I assume this is safe and ok based on all of the numbers.

My bias pot range matches what was brought up earlier. I can get 45-70 from the pot. And I have the correct resistors for a 6550 amp.

 

[harddriver]

I am at 55% of a 6550 so thats 23 watts. I assume this is safe and ok based on all of the numbers.

My bias pot range matches what was brought up earlier. I can get 45-70 from the pot. And I have the correct resistors for a 6550 amp.

When you say you are getting 45-70 from the bias pot is that measured in DCV -45 to -70 on pin 5 of the tube socket? Or are you stating you get 45 to 70 ma when you sweep the pot reading your bias rite readout? They are two different things here so I want to be clear and not confuse the two.

Do you have a multimeter to measure the negative voltage at pin 5? Since you stated in this specific instance we are talking a 100 watt 2203 correct? I know you asked about a 50 watt EL34 amp earlier.

If your amp has the original 150K bias splitter resistors and the factory bias resistors you should be ok to run the 6550's. If you have a quad of tubes that are really really hot on current draw or really really cold on current draw they can be outside of the normal operating bias range but most tubes should be fine.

 

[ericb]

OP, I've used an older version of the Weber Bias Calculator for at least 20 years now. It's a good guideline.. For some reason they list every friggin tube other than 6550's . There's another calculator online which'll suggest 50-58ma of current will be good. 6550's are heartier than KT88's which are heartier than EL34's. They are hi-wattage tubes 42 vs 25ish for El34's, and can take way more current. Good luck. https://www.amplifiedparts.com/tech-articles/tube-amplifier-bias-calculator

p.s my 1980 JMP has about 500 for Plate Voltage and so does my 1971 SuperLead, and if I remember right so does another Metropolis/BFG Superlead I have. . Yes they're out there.

Eric

 

[ericb]

I’ve always read that to be safe, since Marshall transformers are designed for EL34 NOT 6550s, you should bias them as an EL34….so between 30-40ma. If you bias them in the 50+ range like a typical 6550 you’re taking chances on stressing the transformer. The higher plate voltage PTs appeared in 1982, after dropping down in 76/77 in my experience(early 70s Marshalls with the 2562 Dagnall could be 520! My 72 has a 3556 which is 485pv). I’ve experimented with GE 6550s in JMP/JCMs and I didn’t hear a better tone between 35ma or 55ma. So 35 is where they stayed in the last 6550 Marshall I owned. Longer tube life and transformer safety for the win.

Wouldn't his amp have a different Transformer since it was built for 6550's or they used the same? Maybe I'm missing something. My mid 80's JCM 800 2204 was built for 6550's but I had it converted to El34's for cost/ease of getting tubes (30 years ago!) . . . Do they have the same transformers? If not, 6550's can definitely be biased above even 50ma of current at 500plate volts. Curious

Eric

 

[Racerxrated]

Wouldn't his amp have a different Transformer since it was built for 6550's or they used the same? Maybe I'm missing something. My mid 80's JCM 800 2204 was built for 6550's but I had it converted to El34's for cost/ease of getting tubes (30 years ago!) . . . Do they have the same transformers? If not, 6550's can definitely be biased above even 50ma of current at 500plate volts. Curious

Eric

AFAIK Marshall used the same Transformers for all amps; no matter the country they were going to. Unlike Mesa Rectos which had Eu specific PTs. Marks had the Export trans which could be switched to different voltages like Marshalls. Only difference with the 6550 Marshalls is the bias circuit which would be altered to accept 6550s in US amps. Which is why it’s safer(as I’ve read anyway) to bias a 6550 as a 34 instead so not to stress Transformers that were designed for 34s. A true 6550 amp like a VHT Deliverance for example, would rather see a bias set from 55-65ma per tube.

 

[ericb]

AFAIK Marshall used the same Transformers for all amps; no matter the country they were going to. Unlike Mesa Rectos which had Eu specific PTs. Marks had the Export trans which could be switched to different voltages like Marshalls. Only difference with the 6550 Marshalls is the bias circuit which would be altered to accept 6550s in US amps. Which is why it’s safer(as I’ve read anyway) to bias a 6550 as a 34 instead so not to stress Transformers that were designed for 34s. A true 6550 amp like a VHT Deliverance for example, would rather see a bias set from 55-65ma per tube.

Interesting, thanks for the info

Eric

 

[panhead]

AFAIK Marshall used the same Transformers for all amps; no matter the country they were going to. Unlike Mesa Rectos which had Eu specific PTs. Marks had the Export trans which could be switched to different voltages like Marshalls. Only difference with the 6550 Marshalls is the bias circuit which would be altered to accept 6550s in US amps. Which is why it’s safer(as I’ve read anyway) to bias a 6550 as a 34 instead so not to stress Transformers that were designed for 34s. A true 6550 amp like a VHT Deliverance for example, would rather see a bias set from 55-65ma per tube.

Yeah but in the VHT one could set it up to run el34`s. I dont think a transformer is designed specifically for 6550

 

[panhead]

Are we talking output or power? 6550 is akin to 6l6, i had a jcm900mkIII that was designed for 5881`s that someone put el34. The bias would only go to13-15ma, on a whim I put 6550`s in it and it biased rightup. 55ma`s and it was a monster should have kept. Also did the same thing to a friends hotrod 50 replace the 5881 with 6550`s biased up and whole nother level of amplifier

 

[harddriver]

AFAIK Marshall used the same Transformers for all amps; no matter the country they were going to. Unlike Mesa Rectos which had Eu specific PTs. Marks had the Export trans which could be switched to different voltages like Marshalls. Only difference with the 6550 Marshalls is the bias circuit which would be altered to accept 6550s in US amps. Which is why it’s safer(as I’ve read anyway) to bias a 6550 as a 34 instead so not to stress Transformers that were designed for 34s. A true 6550 amp like a VHT Deliverance for example, would rather see a bias set from 55-65ma per tube.

Yes all Marshall Output transformers impedances are specified for EL34 tubes as Racer says, the impedance if the output transformer is a mismatch for 6550 tubes so you either like the sound they produce or you don't, because of that impedance mismatch they were afraid the the 6550 would become unstable and go into runaway bias and then burn up the tubes and output transformer.

The use of the 150K bias splitters mitigated that possibility along with having to increase the bias voltage to get into their operating bias range was Marshalls solution. Actually Marshall hated the way the 6550's made the amps sound but in the mid/late 70's the quality of the EL34 tube supply degraded and Marshall distributors had alot of problems with EL34's getting broken in shipment to the US so Marshall's answer was to use a very heavy duty tube that could survive rough treatment, that was the 6550 which is a beam forming element tube in the same family as the 6L6 and GE/Sylvania 6CA7 which basically are 6L6's with an EL34 pinout.

I read a tube comparsion description that I ten d to agree with and it said KT88's even though they are kinkless tetrode are more akin to a heavy duty EL34 tonewise as they have the mids emphasis(like an EL34) with more output which is why Marshall used them in the Marshall Major 200 watt amps. KT77's are also a very heavy duty EL34 variant.

6550's are a more linear full frequency tube and is basically a larger more heavy duty version of a 6L6. I think they were mostly used in the military and audiophile amplifiers prior to Marshall using them in amp.

 

[Racerxrated]

Yeah but in the VHT one could set it up to run el34`s. I dont think a transformer is designed specifically for 6550

I actually asked VHT in an email if you could run 34s and he said you’d have to change a couple things but the transformers could handle it. I tried KT90s also and they sounded like huge 34s.
Not a tech by any means but the advice I read on another forum long ago is if you have a 6550 Marshall, bias them like an EL34 to avoid any risk of over stressing the OT. I didn’t hear any difference between biasing the GE6550s at 40 vs 55…so I’ve always kept the number lower like a 34

 

[psychodave]

Do you want your bias to drift with shit tubes and run away or not?

My amp has the 150k bias splitters supporting 6550s, but even then, there’s drift until about an hour due to tube quality these days. Actual data sheets show an hour of warmup time and I’ve noticed you need that time else you risk drift.

I use 20 minutes or so... basically I just want to know the tubes are temperature stable.

 

[panhead]

I actually asked VHT in an email if you could run 34s and he said you’d have to change a couple things but the transformers could handle it. I tried KT90s also and they sounded like huge 34s.
Not a tech by any means but the advice I read on another forum long ago is if you have a 6550 Marshall, bias them like an EL34 to avoid any risk of over stressing the OT. I didn’t hear any difference between biasing the GE6550s at 40 vs 55…so I’ve always kept the number lower like a 34

KT90`s are expensive

 

[Racerxrated]

KT90`s are expensive

Yep my pair were 90 shipped back in 15. Afraid to see what they are these days

 

[glpg80]

I use 20 minutes or so... basically I just want to know the tubes are temperature stable.

I don’t have a tube tester like you do ( I know I need one) but this was also to allow drift/burn in on initial fire up.

If it’s just to check or change idle then 30 minutes is plenty for a set of burned in tubes.

New I do burn them in for an hour and I mean that too - I’ve seen drifting trying to cut corners at less time than that. Some companies do 24 hour burn in for comparison for this reason to weed out the duds.

 

[SpiderWars]

It could easily take an hour for those DSLs with the heat-induced bias drift issues to get really hot and it seemed they would just slowly get hotter and hotter. So I believe a full hour could yield better results.

And now that I think about those, I think Marshall recommended to set the bias at like 90mV per side (45mV per tube) on those DSL/TSLs. And those were 475vdc-480vdc with EL34. I remember thinking wow that's really hot.

 

[psychodave]

I don’t have a tube tester like you do ( I know I need one) but this was also to allow drift/burn in on initial fire up.

If it’s just to check or change idle then 30 minutes is plenty for a set of burned in tubes.

New I do burn them in for an hour and I mean that too - I’ve seen drifting trying to cut corners at less time than that. Some companies do 24 hour burn in for comparison for this reason to weed out the duds.

I actually burn them in myself.... even if a tube dealer says they do it. I dont trust any of them.

 

[Cap217]

All good advice. I should burn these tubes in for 8-24 hours and check. I never paid extra for the burn in process but it makes sense now. If they do a 24 hour burn in, then match, that would be great.

 

[glpg80]

All good advice. I should burn these tubes in for 8-24 hours and check. I never paid extra for the burn in process but it makes sense now. If they do a 24 hour burn in, then match, that would be great.

You only really need an hour to 2 hours max. I recommend running them very cold initially to prevent initial red plating. Mine settle after an hour.

By burn in I do mean out of standby and current flowing - so you will have to monitor things. I don’t run a signal through them, usually just idle current.