JCM800 2203kk bias

[punchy712]

I just replaced the tubes in my 2203kk and I was wondering if someone had a somewhat definitive answer on what the bias should be set to? I looked through other threads and it seems like the range is 70-90 mA (which I assume is the same value in mV due to the 1ohm resistor). I'd really like to confirm this though, as other threads mention values much lower, in 30s.

I bought the amp new when it came out and this is the first time changing the tubes, so it's been quite a while. After plugging in the new tubes, the bias came in at around 86/90 per side and I dropped it to 80 for both sides.

 

[V-man]

To those unaware, these are KT-88s

 

[punchy712]

To those unaware, these are KT-88s

Yes thank you, sorry I should have specified that in my post.

 

[TheKman76]

It's possible there is confusion around measuring per valve or across both valves in each pair.

Broadly speaking ~40mA per valve should be a pretty safe bias current, that's around 60% max designed dissipation for typical HT voltages. If the bias measurement resistor is measuring current for both valves you'll want ~80mA for the pair.

Please bear in mind this is ball-park only, do not take this as gospel. To bias the circuit accurately requires knowledge of the exact voltages applied to the valves.

Can you point me to the schematic you're referring to for this model?

 

[FourT6and2]

You need the exact plate voltage in your specific amp. Nobody can just pull a number out of their bum. Rule of thumb is you want bias hot enough to where amp is out of crossover distortion, but don't exceed 70% of max plate dissipation for the tube you're using. So basically... set the bias to where it sounds good.

 

[punchy712]

Thanks a lot for both answers. I didn’t even clue in that some people may be posting the bias of an individual tube or a pair, as it is measured using the built in probe. Sorry I don’t have the schematic handy and I don’t seem to find a link to it anymore, but I’m using the built in two probes to measure, which I believe is the bias for each pair.

FourT6and2, if you don’t mind me asking for some more details on how to calculate the target bias, and really I’m looking to make sure I don’t exceed a safe value, otherwise I’ll adjust based on the sound as well.

What value should I use for the max plate dissipation, or is this something that I need to find from the spec of the particular tube I’m using? Also, to calculate plate voltage is it at pin 3 of the tube (which would be the third pin clockwise as looking from below)?

 

[punchy712]

Attached is the schematic I was using.

 

[Harlequin tusk]

Bias Calculator – Weber Speakers

www.tedweber.com

Go to this site and enter your info

 

[Pete Farrington]

I suggest to regard KT88 as having a 30 watt anode dissipation limit.
Based on it probably being what it would be under the design centre rating system.

Its design maximum limit is 35W, absolute maximum limit is 42W.

If we regard EL34 as having a 25W limit, that’s under the design centre system, so it seems appropriate to regard kt88 as having a bit lower than 35W.

So, I suggest to idle at not above 21W (70% of 30) per KT88.

Assuming a HT of about 450V, that works out to 47mA anode current per KT88.

 

[punchy712]

Thank you Pete for that great breakdown, I appreciate all the details. The important bit that’s going to be more specific to my amp is the HT/plate voltage. To confirm, do I measure this at pin 3 of one of the power tubes?

 

[Pete Farrington]

do I measure this at pin 3 of one of the power tubes?

I prefer to measure the HT voltage somewhere convenient, as it’s less problematic than an anode lug. But I don’t know your amp’s internal layout or even schematic, so can’t tell you exactly where to check the HT.

If you’ve not got a bias probe which facilitates anode voltage measurement, I suggest to use a probe clip on any if the KT88 socket pin 3 lugs.
Clip it on with the amp de-energised and unplugged from the wall outlet. Then power up, and let the amp idle (ie not in standby mode) for a bit before taking a measurement.

Keep the probe cable away from any preamp controls or circuitry.

Using a regular probe tip has the risk of it slipping off the lug and shorting things out.

 

[punchy712]

Thanks again Pete, I really appreciate your explanations and walking me through this as I understand better now how to calculate it.

I measured HT at the transformer and it was 450V off standby (490V on standby) so your calculations are spot on. I previously had the bias set to 43 mA per tube which is close to your value so I’ll leave it as is for now as the tubes break in.

 

[TheKman76]

EDIT: I missed the bit that said 43mA *per tube*... whole post of nonsense.

Interesting @Pete Farrington that you and I target the same power with different approaches. I target 60% of design max, or 21W, you target 70% of a down-rated design max, 21W.

 

[mickeydg5]

Marshall basically designs its amplifiers to run 6550/KT88/KT90 at a bias of 32mV(mA) - 47mv(mA).
This has been discussed in many threads especially 2203KK amplifier threads.
Most people have tended to hover around 40mV(mA) - 42mV(mA) with the 2203KK.

These amplifiers are biased as 100 Watt output no matter what output tube is loaded. So no should bias it as if maxing out a 6550/KT88 type at 30W or 35W dissipation. The numbers for 6CA7/EL34/KT77 should be used as a maximum for bias being at about 25W dissipation because that is how the power amplifier is designed.

REMEMBER, the 2203KK like many other Marshall amplifiers use the 1-ohm bias monitoring resistor method WHICH INCLUDES the screen current along with the plate current. So you have to add the calculated plate current and the calculated screen current. For these output tubes most just average idle bias screen current at about 4mA - 8mA.

Use lower bias values with higher HT voltages.

So:
25W / 450V = 56mA
56mA x 70% = 39mA
39mA + 8mA = 47mA
Therefore, 47mA is a maximum idle bias per tube with average to lower HT voltages when using the 1-ohm bias monitoring resistor method.

Hopefully this helps explains these tubes and idle bias in amplifiers like this.