1977 Marshall 2203 High Plate Voltage!!

[Chuck H]

hence this might be why I can't get lower than 41MVDC per tube.

This is confusing me. Where is the 41mV DC? Do you mean mA current? You have less than that now.

[dstrat]

Thanks for the response. One thing I did notice on the board (bias section) is that this amp has a 65k bias range resistor instead of the typical 47k in that spot. hence this might be why I can't get lower than 41MVDC per tube. Possible that this amp's bias section wasn't properly converted to EL34? here is a photo of that section.

[ATTACH=CONFIG]47909[/ATTACH]

Looks like a 56k to me, unless im going colorblind.

[cluster]

This is confusing me. Where is the 41mV DC? Do you mean mA current? You have less than that now.

I installed a 1 ohm resistor to ground on the 1st power tube so as to conveniently measure the current. So yes, it should read as 41 mA.

[cluster]

Looks like a 56k to me, unless im going colorblind.

Yup,you are right. Sorry, my bad. But shouldn’t it be 47k as per schematic??

[cluster]

So, given that I have installed a 1 ohm bias sense resistor on the 1st power tube and that at 530vdc I’m seeing 41mA of current at that tube and don’t have the range to bring it any lower, should I swap out the bias range resistor? I may consider putting another one ohm bias resistor on the last power tube on other side of the OT primary.

[dstrat]

schematic says 56k
usa 47k
guess you could try 47k see if you get more swing
also r8 should be 15k with the 47k for USA models.

edit also you should be shooting for around 30ma for one tube or 60 for two .. guess you are talking 2203

[dstrat]

my eyes deceive me I thought r27 was 10k but it might be 15k sorry

[cluster]

yes... it's a 2203. with my current bias circuit I can't get enough negative grid voltage to drop the current at the tubes below 41mA. also, for got to mention that bias feeds are 100k instead of the 220k (6550 vs el34). Would anyone recommend a piggy-backing a resistor on the 56k (parallel) to see if that helps?

[dstrat]

i was going to suggest just that but want to ask what is the range you get from bias voltage?

[dstrat]

ok found it in previous post -51v you need to get to at least -56v or so change the r8 before the diode to a 22k or even 15k should give you a better range to get the current down to around 30ma or so.

[pdf64]

I find *BOTH* primary side current methods DANGEROUS, shunt or not.

Safe method is to "waste" 2 minutes and add 1 ohm cathode resistors, then measure voltage drop across them.

Added advantage: Math is easier !!!!

Another benefit is that with a 4 tube amp, it's possible to see what each tube is doing, rather than just assuming that the parallel pairs share the current (ie in each leg of the OT primary) equally; as it often turns out not to be a 'safe' assumption

[dstrat]

agree ^^^^
I am also wondering if this amp bound for north america did have 6550s installed?

[cluster]

Next step. I added a 120k in parallel to the 27k bias range resistor so as to get my bias in the proper range. So now I was able to get the tubes to just about 30mA each. Plates still high at 538VDC. Grid bias -50.6VDC.

Also measured the voltage drop and resistance across the OT/CT and got:

0.83VDC and 1.01VDC
15.4 ohms and 14.8 respectively (power off)

or 5.4 mA and 6.8 mA (maybe the difference is because of the tubes not matched)

...so using the 70% of the maximum plate dissipation of the power tube EL34's plate dissipation is 25 Watts maximum. Then 25W times 70% equals 17.5W

17.5W/538VDC = 0.032A per tube or 32mAmps/tube

Thoughts?? Also, anyone recommend the 220k grid resistors vs the 100k that are there now. EL34 calls for 220k. 6550 calls for 100k. Tone Lizard recommends 100k for EL34 for longer tube life and darker sound.

[dstrat]

I'd say if it sounds good, it is.
100k will help with tube life , they say.

[g1]

I am also wondering if this amp bound for north america did have 6550s installed?

It wasn't a North America thing, strictly USA.
This being a Canadian (CSA) model would have had all the British (non US) value resistors and come with EL34's. Some other strictly CSA 'features' for various models included a whole extra board of fuses, the metal toggle switches rather than plastic rockers, and lack of 16ohm tap (too high voltage accessible to user).

In this particular amp, I think someone changing the 220K grid resistors to 100K would have tweaked the bias range a bit, and the higher plate voltage further compounded the problem. Sounds like you should be good now.