Preamp supply sag

[g1]

Are you measuring AC? DC? Do you have a true RMS meter?
Can you post the schematic of the circuit, it is not possible to tell from the thread which actual circuit is being discussed here anymore.

[carlg]

Yeah that sounds like a measuring problem. In the location you have it, the resistor wouldn't be expected to have a DC voltage across it. The diodes are off most of the time. They only open for a short time at the top of their cycle to slurp a quick pulse of electrons off the filter cap. That short current pulse will be higher than the average current draw of your amp, but then the current falls to zero until the next pulse comes through.

It's very much like bailing out a boat with a slow leak. The boat is your filter cap, you want it to stay well drained of "water" (electrons). The slow leak is your power amp, it's pouring a continuous flow of electrons into your cap. The power supply diodes are the bucket. Every now and then they take a big scoop out of the water in the boat. That keeps the average level of water in the boat down, but they're not a continuous flow like a pump would be, only sudden scoops. Because of that there is no simple DC voltage across a resistor that's before the filter cap. Some meters can average all those current pulses out and tell you what an equivalent steady state voltage or current would be, some can't. It's not terribly helpful to you here even if it can because that average value isn't what's actually taking place.

[EFK]

Okee dokee, since I don't have any single 600V rated electrolytics here, I quick twisted up (4) 1uF/630 V plastic caps in parallel for a 4uF total, which ought to be enough just for testing purposes. I attached this *before* the 450 ohm resistor. Basically the same results (voltages a tiny bit higher but my wall voltage is higher today), the resistor is only dropping around 4-ish volts at idle and 8-ish when the amp is hit with signal and maxed. The resistor doesn't even get warm.

I don't have a schematic because this is a prototype project I've been working on and I've been winging it as I go along. I'm using a Fluke true RMS meter for measurements.

PT is 190/0/190, w/ a 6CA4 tube rectifier. Off each secondary is also a reversed diode to ground, and I'm using the center tap so I've essentially made a voltage doubler off the tube rec. The high (@ 500vdc) dc is running into a totem pole setup, currently dual 16uF cap cans each paralleled and in series w/ 100K sharing resistors. Preamp is running off the top of the totem, 500 vdc through an extra RC filter to drop and clean up the voltage feeding the ltpi and then on through the preamp. The PT center tap is running to the junction of the cap cans in series, and this is where I'm pulling my power amp dc (@250vdc) for plates and screens. the power tubes are running a total of @ 70 mA at idle, climb to @ 110-120 mA under full load. So, I would expect a 50 ohm resistor to initiate quite a bit of voltage sag. It isn't.

I posted a link above to the rectification setup I'm using, I'll repost it here. When you follow the link, I am using a setup just like the 4th drawing down from the top of page, except I am using only a single totem/series cap filter, not two pairs (drawing shows two totem pole filters).

Link: http://www.tubecad.com/2012/05/blog0230.htm

I'm thinking the only thing left for me to do is to test my theory that I really need to split up the dropping resistor into three separate resistors to 'balance' out this setup: one off the tube rec, one between the reversed diodes and ground and one on the PT center tap. I'm pretty much out of ideas at this point, other than to try again with an even smaller and saggier PT. Have to say though, this one is not oversized to begin with.

[EFK]

Ooooh, took a gamble today and it definitely paid off. Two words - mixed bias. I know some people feel it doesn't make much of an audible difference but in this case it sure does. Very noticeable. Wanted to maintain a nice even number for easy bias readings so I popped a 100 ohm resistor on each cathode, *unbypassed* (I prefer an unbypassed cathode R in most amps). This seems a good balance and it is definitely adding some compression. I think I've got it set now and (for today anyway) I'm getting what I want out of it. It also sounds better as the amp transitions from clean to dirt, definitely smoother.

[g1]

Just one question on something I missed. Your OT gets it's feed after your sag resistor, or before?
From the following statement I get the impression the OT is before the sag resistor, right off the rectifier (so no sag provided by power tubes):

The PT center tap is running to the junction of the cap cans in series, and this is where I'm pulling my power amp dc (@250vdc) for plates and screens. the power tubes are running a total of @ 70 mA at idle, climb to @ 110-120 mA under full load. So, I would expect a 50 ohm resistor to initiate quite a bit of voltage sag. It isn't.

This is why it's so hard to discuss without a complete schematic.

[EFK]

Here is quick sketch of the relevant portions; I have the output sockets wired to handle the 4 output tube types listed because I have hundreds of them! I generally have been using 6CM6 which are a 6V6 in a 9 pin bottle, and very tough. Measured at the cathodes, my output is pulling about 70 mA at idle and 110-120 mA under full load, so I would anticipate more than only @ 8 V of sag from idle to full power even without a sag resistor. Anyway, on this drawing there are three points marked with an "X," numbered 1, 2 and 3. I have tried up to a 450 ohm sag resistor on points one and three because my goal has been to get the entire PS to sag and I *think* that putting a resistor only at point 2 would cause the output to sag, but not the preamp. Putting a single resistor at points one or three makes no difference, I get the same minimal voltage sag as if the resistor is not even present. What I have not tried is putting a set of three smaller sag resistors, say 100 ohm each, at all three points simultaneously.



I have the dropper to the preamp on a switch so I can toggle between earlier JTM45/black flag 50 preamp voltages and later metal panel voltages.