JVM 205 Bias Issue

[jojobea]

Long time lurker, first time poster.

I have a 2013 JVM 205H that I'm having a few issues with.

As I think it might help diagnose it this is what happened in order:

• I bought this amp used (a while ago) and noticed recently that it had the original power tubes in it
• I ordered some new ones and in the meantime learned a bit about biasing to see if I could do it myself.
• With the old tubes the plate voltage was around 433V, and each of the tubes was just 8mA and 10mA respectively.
• Everything I'd read suggested these should be around 65-70% dissipation which is high 30's mA given that plate voltage.
• I adjusted the bias to just 30 each side as I figured, hey these are very old tubes, let's just put it back to what I'd read is close to the factory setting.
• Within 20 minutes or so both tubes started red plating. I dropped the mA to 20ish and it seemed to stop the red plating temporarily but actually within an hour of playing they were red plating again.
• So I thought, these are 8 year old tubes, let's try the new ones first before panicking.
• Today I've got the new ones and installed them. When I measured the bias the plate voltage was now 443V and the tubes were around 20mA each.
• I then tried to adjust them up and the max I could get them to before hitting the max turn on the pot was around 30mA. I thought that's a bit odd but let's hear it. They weren't red plating so I figured it's all good to at least see what it sounds like.
• The sound was pretty terrible. Not obviously broken terrible, just loose, noisy, a bit harsh. The new tubes are JJ EL34's and generally my experience with JJ's is they're dark and perhaps a bit bland but it didn't sound right.
• So I dropped the bias down to 15mA each side just to see what it sounded like. It cleaned it up and was definitely better. The red plating is gone regardless (on both the 30mA and the 15mA).

So I'm at a loss to understand why the bias setting is maxing out at 30mA and why 30 sounds harsh and noisy. With the bias set at 15 or so it sounds 'pretty good'. It may even sound great after a few hours of loud playing which I'll do on the weekend. But it still doesn't seem right to me.

I'm willing to take it to a tech but unfortunately looking around they're all on holidays until Jan 10 or later.

Any recommendations or suggestions would be great. With the old tubes I could get to 35-40mA without maxing the pots out. With the new tubes, 30 is the limit and it doesn't sound good. What could be the issue?

EDIT: The original tubes are Marshall branded VLVE-00010 which according to this forum is Svetlana. I found a post in another forum where someone was describing svetlana el34's being much hotter than standard JJ El34's. Could this simply be that the standard JJ is not a good match and if I don't want to dick around with resisters I should consider a different set of EL34's?

 

[Pete Farrington]

Dunno anything about your amp, but how exactly are you measuring the currents that you’re citing? As in step by step detail.
Because whilst it’s easy to do, it’s also easy to screw it up, take the measurements incorrectly, and come away with nonsensical readings.

https://www.drtube.nl/schematics/marshall/jvm205-65-02-iss3.pdf

 

[jojobea]

Dunno anything about your amp, but how exactly are you measuring the currents that you’re citing? As in step by step detail.
Because whilst it’s easy to do, it’s also easy to screw it up, take the measurements incorrectly, and come away with nonsensical readings.

Hi Pete. For the plate voltage I'm letting the amp warm up on standby for about 15 minutes before coming off standby into full on position, then grounding against the chassis while probing pin 3 on each power tube. For the current I'm using the JVM's 3 pin test connector, grounding on the centre COM pin and using the left and right test pins to measure each side (50 watt amp so only 1 tube per test point).

From what I've gathered, the JVM is one of the easier amps to measure and bias. My multimeter is a good one, definitely not a cheapy. I'm using alligator clips for all connections and the readings while odd to me have been consistent.

 

[Pete Farrington]

For the current I'm using the JVM's 3 pin test connector, grounding on the centre COM pin and using the left and right test pins to measure each side

How exactly are you setting the meter? ie it needs to be set for reading mV, not mA.

 

[jojobea]

How exactly are you setting the meter? ie it needs to be set for reading mV, not mA.

Well, clearly I'm an idiot and that's exactly what I was doing wrong. Let my stupidity be here forever for all to see so others figuring this out don't make the same mistake.

So unfortunately I'd watched a lot of videos and often people were measuring mA directly, however not necessarily using test points (depending on the amp). So I think because people refer to mA in their bias values and some were measuring mA I didn't realise that I needed to measure mV on the those test points on the JVM.

Thanks for your input. The good thing is, the ear test got me to around 39mv on the test point as the sweet spot. I was measuring it wrong but had set the bias about right anyway. Thanks again.

 

[Pete Farrington]

Haha, don’t worry about it, it’s an easy and perfectly understandable mistake to make. And one you’re not going to make again