Tweet
Let me preface this right off the bat by saying I have no scope. I am scopeless. I am deficient in scope capability.
I have an amp that is a mostly Marshall preamp into a semi-Marshall PI (bigger tail resistor, both 100K plate R) - kind of like MArshall meets Vox PI. The power section is a quad of 7591 that are cathode biased, and I'm running them at fairly low voltage with a larger screen dropper for sag and so I can bias it pretty hot at idle without much cathode shift under load.
When I inject a 1 khz signal into the input, and using a resistive load, I am getting a weird situation that I can;t seem to straighten out. If I measure the AC voltage at the grids of the inverted pair of tubes, it climbs as expected up to about halfway on the volume knob, then drops back quite a bit. Consequently the voltage dropped across the cathode R (shared per pair) bounces around a little but stays pretty static, actually dropping a bit. The AC at the grids of the non-inverted pair, meanwhile, matches the other side pretty closely up through the halfway volume point, but then continues to climb and meanwhile so does the cathode voltage. There is a pretty big disparity between the two sides - the non-inverting pair is being hit at max volume with about 12 to 13 VAC, while the inverting side climbs to about 7 VAC at half volume (they both do) but then drops back to around 5 at max. That's more than twice the AC being sent to one side only. ??? I have moved wires around all over the place, relocated the grid wires from the PI in two or three different ways and now I'm using belden shielded cable, with no difference. I have pretty good sized grid stoppers on there too, 10K per tube. Adjusting the PI tail resistor up or down a bit (currently it is 27K and a 4.7K final, with the NFB in between) doesn't make a different. Disconnecting the NFB makes no difference.
I don't really know where to go from here at this point. I suspect there is some kind of imbalance going on in the PI, or a parasitic, but I can;t seem to come up with anything that makes any difference. I have swapped PI tubes to no avail, I have moved the power tubes around and I have alternately reversed the grid leads off the PI, and it's always the same: the non-inverting pair climbs, the inverting pair sags.
Any ideas? Thanks.
I have an amp that is a mostly Marshall preamp into a semi-Marshall PI (bigger tail resistor, both 100K plate R) - kind of like MArshall meets Vox PI. The power section is a quad of 7591 that are cathode biased, and I'm running them at fairly low voltage with a larger screen dropper for sag and so I can bias it pretty hot at idle without much cathode shift under load.
When I inject a 1 khz signal into the input, and using a resistive load, I am getting a weird situation that I can;t seem to straighten out. If I measure the AC voltage at the grids of the inverted pair of tubes, it climbs as expected up to about halfway on the volume knob, then drops back quite a bit. Consequently the voltage dropped across the cathode R (shared per pair) bounces around a little but stays pretty static, actually dropping a bit. The AC at the grids of the non-inverted pair, meanwhile, matches the other side pretty closely up through the halfway volume point, but then continues to climb and meanwhile so does the cathode voltage. There is a pretty big disparity between the two sides - the non-inverting pair is being hit at max volume with about 12 to 13 VAC, while the inverting side climbs to about 7 VAC at half volume (they both do) but then drops back to around 5 at max. That's more than twice the AC being sent to one side only. ??? I have moved wires around all over the place, relocated the grid wires from the PI in two or three different ways and now I'm using belden shielded cable, with no difference. I have pretty good sized grid stoppers on there too, 10K per tube. Adjusting the PI tail resistor up or down a bit (currently it is 27K and a 4.7K final, with the NFB in between) doesn't make a different. Disconnecting the NFB makes no difference.
I don't really know where to go from here at this point. I suspect there is some kind of imbalance going on in the PI, or a parasitic, but I can;t seem to come up with anything that makes any difference. I have swapped PI tubes to no avail, I have moved the power tubes around and I have alternately reversed the grid leads off the PI, and it's always the same: the non-inverting pair climbs, the inverting pair sags.
Any ideas? Thanks.
Comment