I am looking at a vintage 5E3 and noticed somebody had placed a 680k resistor in parallel with the 56k plate load resistor effectively reducing it to 52k. I thought it was there originally because the plate load resistor had drifted higher but it hasn't. The plate load resistor and the cathode resistor are both closely matched which I thought is a good thing but for some reason somebody had purposely mismatched the loads.
I have seen this done with the LTP but not a split load. What effect does this have?
On a side note this amp was modded with negative feedback, resistors added to separate channels, and 47R screen resistors on the 6v6's. I usually see 470R screen resistors or at least 100R. I will probably just return it to stock condition except for the screen resistors (maybe I will increase the value to 470R as I usually do). The tubes are running way too hot so I am planning on increasing shared bias resistor to 330R. Another mod I like because the owner will be bringing this into distortion territory is add a large grid stopper to the grid of the PI.
I have seen this done with the LTP but not a split load. What effect does this have?
On a side note this amp was modded with negative feedback, resistors added to separate channels, and 47R screen resistors on the 6v6's. I usually see 470R screen resistors or at least 100R. I will probably just return it to stock condition except for the screen resistors (maybe I will increase the value to 470R as I usually do). The tubes are running way too hot so I am planning on increasing shared bias resistor to 330R. Another mod I like because the owner will be bringing this into distortion territory is add a large grid stopper to the grid of the PI.
Presumably it's done to increase distortion?, unbalancing the PI seems a common technique in guitar amps.
So it is the same idea as using two different plate loads in the LTP PI, interesting.
I personally am a fan of the sonic attributes of the 5E3 circuit and before opening this up I knew it didn't play/feel/sound like a true 5E3. My fellow band mate who owns the amp never uses it and noticed my 5E3 clone I built sounded different. His tech that did the cap job on it years ago must have did the modifications which are:
1) 12AX7 for the first gain stage instead of the 12AY7
2) 35db of global feedback fed into the second gain stage's cathode
3) purposely mismatched loads on the PI
4) "mic" and "instrument" channels no longer interactive
5) probably not done by the tech but it has an unknown ceramic magnet speaker.
After the cap job and mods were done I am surprised that the amp went out with the output valves operating at 150% rated plate dissipation but maybe that's designed for job security ;-)
End of rant
I personally am a fan of the sonic attributes of the 5E3 circuit and before opening this up I knew it didn't play/feel/sound like a true 5E3. My fellow band mate who owns the amp never uses it and noticed my 5E3 clone I built sounded different. His tech that did the cap job on it years ago must have did the modifications which are:
1) 12AX7 for the first gain stage instead of the 12AY7
2) 35db of global feedback fed into the second gain stage's cathode
3) purposely mismatched loads on the PI
4) "mic" and "instrument" channels no longer interactive
5) probably not done by the tech but it has an unknown ceramic magnet speaker.
After the cap job and mods were done I am surprised that the amp went out with the output valves operating at 150% rated plate dissipation but maybe that's designed for job security ;-)
End of rant
Well, I think the mis-matched plate loads in the stand LTPI circuits we see are actually an attempt to improve balance, not disrupt it. This is what I tend to go by:
Balancing Long-Tailed-Pair Phase Inverter Gains
Now, I wonder if the 1500 bias resistor on the cathode of the 5E3 splitter causes the output on the cathode side to drop by a few %, and that the reduction of the plate resistor was to match the p-p voltage to the cathode. Since the current it determined by the cathode circuit 🙂confused🙂 then a smaller plate resistor would yield a smaller signal…
Or maybe not!
Brian
This makes more sense! And follows suit with the other mods to clean up the amp like negative feedback.
I remember reading about this when I built myself some Mullard 5-20 monoblocks. This is a quote from the book: Tube Circuits for Audio Amplifiers
"The anode load resistors R11 and R12 (R12 is on the side of the LTP where the grid is grounded through a cap) should be matched to within 5%, R12 having the higher value for optimum operation. Optimum balance is obtained when the effective anode loads differ by 3%."
I remember doing this and I did have better balancing with the results.
You are all bouncing around the issue, but not pinning it down precisely.
The AC balance of a split load stage is determined by the matching of the anode and cathode loads. In the case of the 5E3, the 1500 ohm bias resistor that is part of the bootstrapping circuit unbalances the stage since anode and cathode resistors are both 56K, good enough for a commercial guitar amp. It looks like you have a 5E3 where someone couldn't accept this and put a 680K in parallel to geth the anode and cathode loads more equal. I doubt it was done as part an elaborate, measured exercise.
The AC balance of a split load stage is determined by the matching of the anode and cathode loads. In the case of the 5E3, the 1500 ohm bias resistor that is part of the bootstrapping circuit unbalances the stage since anode and cathode resistors are both 56K, good enough for a commercial guitar amp. It looks like you have a 5E3 where someone couldn't accept this and put a 680K in parallel to geth the anode and cathode loads more equal. I doubt it was done as part an elaborate, measured exercise.
exactly as you said it, and also the person who did that didnt know that the split load inverter is operating under heavy local feedback, largely csncelling
imbalances in the load resistors. Better return it to stock, OP....
But doesn't the internal feedback only force the cathode to follow the grid very closely? What feedback is there from the anode? I mean, if the cathode follows the grid almost perfectly before clipping, and the grid cathode voltage determines anode-cathode current, and the same current must flow through anode as the cathode, then how could two different resistor values yield the same p-p signal and not violate Ohms law?
Yes, but do not forget the feedback from grid to anode, too. The cathode to grid feedback is applied to the whole stage, the anode is not separate from that. Yes. there will be a difference in the outputs, but not nearly as much as you would expect, because the feedback fights against that.
- Status
- Not open for further replies.