Wonderland/JM100 build

[pompeiisneaks]

I just wanted to make one note about voltages. Realize that each tube added drops the voltage a bit more, as they draw current. So you likely already saw voltage drops with the power tubes in, as you add the preamp tubes it will drop as well. You could compare to some other existing amps voltages and see if the power amp stage with current draw is near yours, then you'll likely see the same voltages at the preamp stages. Just something to help you realize the unloaded voltages are quite high, but come down with tubes.

~Phil

[martin manning]

...can i ask; if the transformers came out of a working amp how did peavey get enough negative voltage to BIAS?

Good question. It looked tight from the start. I'll sketch something up after I get home.

[ToneMerc]

...can i ask; if the transformers came out of a working amp how did peavey get enough negative voltage to BIAS?

Good question. It looked tight from the start. I'll sketch something up after I get home.

The PV config uses the entire winding of 23-0-23 correct whereas, Marcus is using 23-0.

TM

[martin manning]

I don't think Marcus is using the bias winding CT. He measured 45 VAC earlier, which would yield ~60VDC... that may not be enough. Here's a doubler supply. Adjust the input resistor to get the bias range you need, but 1k8 @1W should be close.

[norburybrook]

thanks Martin.

I am using the two 23-23 to get the 46v which as MArtin said isn't enough.

OK what value cap? I don't think I've got another 100uf 100v

give me a range of values that will work, obviously as long as the voltage is above 100 it's fine.


regarding the resistor on the board. 1R is very small so what sort of range would you guess would be a good starting point?

thanks again..that list of beers is growing longer Martin, if we ever meet you'll be pissed as a fart as we say in the UK

Marcus

[martin manning]

Adding another 100u would be ideal, but in any case the capacitances should be matched. If you have a couple of 47u you could parallel them to make another 100u, or replace the one you have with a 47u and add the second one at 47u. I agree they should be 100V rated. You could go lower in capacitance (a pair of 33u or 22u), but at some point you will get some hum in the output. For the resistor you'll probably have to experiment. I'm guessing it would be in the hundreds of ohms, and I think you'll want at least a 1W.

I'll never turn down a pint offered in appreciation, but I'm a quality, not quantity kind of guy ;^)

[ToneMerc]

I don't think Marcus is using the bias winding CT. He measured 45 VAC earlier, which would yield ~60VDC... that may not be enough. Here's a doubler supply. Adjust the 1R to get the bias range you need.

Martin, shouldn't the bias winding have a reference to ground with at least the CT or the opposite 23V lead grounded? I don't see how connecting the 23-23 leads together as a single tap as working correctly. What am I missing?

On the surface 46VAC should be enough as there were Fender PT's with 45 and 50 volt bias taps that biased 6l6's.

TM

[martin manning]

The PT has a separate bias winding at 23-0-23. The original schematic shows -56V bias with 500V on plates and screens, which is what Marcus is seeing. Most Fenders did not run voltages that high, and some tubes will need more negative voltage to bias. Peavey was cutting it too close IMO.

[norburybrook]

I don't think Marcus is using the bias winding CT. He measured 45 VAC earlier, which would yield ~60VDC... that may not be enough. Here's a doubler supply. Adjust the 1R to get the bias range you need.

Martin, shouldn't the bias winding have a reference to ground with at least the CT or the opposite 23V lead grounded? I don't see how connecting the 23-23 leads together as a single tap as working correctly. What am I missing?

On the surface 46VAC should be enough as there were Fender PT's with 45 and 50 volt bias taps that biased 6l6's.

TM

they're not connected together, one side of the 23-0-23 goes to ground and the other side goes to the rectifier board to give 46vAC and the CT is tied off. Rectified it gives about -60vDC which isn't enough to bias at 520VDC on the 6L6 plates.


Marcus

[martin manning]

Hold on Marcus, I need to correct my drawing... Ok, edited in above. Just connect the second diode to the positive end of the second cap rather than the negative, and the other AC goes to the junction of the two caps, not to ground.

[norburybrook]

thanks Martin,

I'd finished for the day anyway, the time difference

I'll take the board out today and Rog B from the forum is in the studio with me tomorrow doing some recording and he's got a spare 100uf 100v cap he's bringing over.

just want to clear up the technical side of this in my head.

I would have thought that without enough voltage from the bias my output tubes would be red plating, no? or is it the other way round. Looking at them yesterday they were no where near red plating visually so that confused me, along with conceptually grasping the correlation between the -ve voltage from the bias and the mv reading on the tube pin. I had inadvertently had the pot at the wrong end of the range when seeing -58v originally so when I started at -77mv and turned the pot it went upwards past 100mv rather than down towards the 39 needed however the plates seemed fine.

Is the 1.8k a rough estimate for resistance value? I don't think I've got anything like that I know I have some 470R and 100R , you did mention a range in the 'hundreds' originally.

(edit..found a 1w 1.2k I'll start with that)

And......

do I need to connect everything up in the amp and test the bias range when checking this, or can I quickly check the voltage somehow to see if i'm in the range?

Marcus

[martin manning]

100% Pa at 500V would be 60mA so you weren't too far over, and it would probably take some minutes before everything got hot enough for you to see the plate glow.

If the lowest voltage reading you could get across a 1 ohm current sense resistor was 77mV, then you need to get lower (more negative) bias voltage.

The 1k8 value is from modeling, and you should get a range of ~60V-80V. That is in the hundreds... 18 hundred ;^) The current through it is pulsating DC so its not a simple calculation.

You can use a half-watt resistor there so you are likely to have something close, or parallel a couple of values to get 1k8 or 2k.

To test it you'll have to connect everything up, including the 10k pot with its 27k to ground.

[norburybrook]

thanks Martin.

I've drilled the board and put the eyelets in so just waiting for RogB to bring the 100uF/100v tomorrow.I'll probably get to it tomorrow evening after my studio day is over.

it's been a journey this one so far, but a great learning experience again.

I'm hoping this 'mature USA Iron' will be worth all this faffing around to get it to work.



Right back to work, got a day of Bass and guitar sessions ahead of me.

have a good day, as you say over there actually it's become part of UK parlance now.


Marcus

[rogb]

Along with the 100uF, I've got a 2.1K 2w MOx and a RN65 10k that will parallel to 1.8K.
Or use the 2.1K on its own.

[martin manning]

The 2k1 by itself should work. Looking forward to the report.