the_manta said:When you accept that,
-you need output 2 trannies per channel (doubles the weight & cost)
-DC will NOT cancel anymore (need of large core & airgap)
, then there's no reason why you shouldn't do it
And the power won't be the square of the tubes, minus losses anymore ... they'll just be paralleled SE tubes.
PeteN said:
This really only matters if operating class B, I doubt it matters at all with what I wil call "summed anti-phase SE" for the lack of a better term. I think about power will be about equivalent to a pair of SE in parallel, possible the even harmonic cancellation of PP would be retained.. (I haven't thought about this.)
I absolutely agree with Kevin in that the even harmonics will still sum and by nature of them being out of phase they will appear to cancel. As he mentioned this is only the case for pure class A operation and once you near class B the coupling of the windings via core and proximity becomes a much different situation. That said, it only makes sense to me that a transformer wound for class A has different needs than one wound for AB just as a transformer wound for a triode has different needs than one wound for a pentode.
I say go for it and report back. If you already have the parts on hand in a junkbox what do you have to lose? If you purchase them and the project is deemed a failure, you now have a really cool junkbox for future builds.
dave
I say go for it and report back. If you already have the parts on hand in a junkbox what do you have to lose? If you purchase them and the project is deemed a failure, you now have a really cool junkbox for future builds.
dave
The issue is that I built this circuit ...
... and it doesn't work at all. Total floparoo. Well, it amplifies the signal, but the noise is louder than the output. I was super careful with grounding, the heaters are regulated DC, the B+ supply's ripple is in the uV range before the CCSs, and the PS is a long way away. I also tried lots of different tubes, but no improvement at all.
So, I thought I'd back up a step. I have built lots of single ended parafeed linestages. Now the plan is to buid two of them, get them working individually, then put them together to see if I can track down the issue, then swap in the single output transformer -- unless someone sees an error in the schematic.
An externally hosted image should be here but it was not working when we last tested it.
... and it doesn't work at all. Total floparoo. Well, it amplifies the signal, but the noise is louder than the output. I was super careful with grounding, the heaters are regulated DC, the B+ supply's ripple is in the uV range before the CCSs, and the PS is a long way away. I also tried lots of different tubes, but no improvement at all.
So, I thought I'd back up a step. I have built lots of single ended parafeed linestages. Now the plan is to buid two of them, get them working individually, then put them together to see if I can track down the issue, then swap in the single output transformer -- unless someone sees an error in the schematic.
Hi Doug,
What kind of noise? I don't see anything inherently wrong with this design on a quick inspection. Are you sure something is not oscillating- often vhf oscillations will manifest as odd hums and buzzes.. (and other things) Could be the ccs oscillating too.
You mention the power supply is a long, long way away - any local supply decoupling? You might be surprised at what the supply side of a ccs might do if presented with something other than a low rf source impedance..
What kind of noise? I don't see anything inherently wrong with this design on a quick inspection. Are you sure something is not oscillating- often vhf oscillations will manifest as odd hums and buzzes.. (and other things) Could be the ccs oscillating too.
You mention the power supply is a long, long way away - any local supply decoupling? You might be surprised at what the supply side of a ccs might do if presented with something other than a low rf source impedance..
It is like a mix of loud hum and other random noise -- that is, not a perfect sin wave type hum. I once had a ground trace burn up on a power supply PCB (which led to the ground having a high resistance) and I got a similar noise, but the ground all looks pretty good here. I also got a similar noise when I tried to build a 01A parafeed linestage, but that was a long time ago and I attributed it to a noisy filament supply
There is a CLC supply (I also tried a CRCRCRC supply I found in a drawer) about 18 inches away, and then a RC (50mF oil cap) about 3 inches from the CCS. But, there is no local decoupling -- maybe I'll try a 0.1uF cap closer to the CCS and see if that helps.
There is one thing that bothers me about the build. I used this schematic for the volume control and input transformer, etc. The leads from the tubes to the pot are quite long, though I moved them around and it didn't change the noise at all.
There is a CLC supply (I also tried a CRCRCRC supply I found in a drawer) about 18 inches away, and then a RC (50mF oil cap) about 3 inches from the CCS. But, there is no local decoupling -- maybe I'll try a 0.1uF cap closer to the CCS and see if that helps.
There is one thing that bothers me about the build. I used this schematic for the volume control and input transformer, etc. The leads from the tubes to the pot are quite long, though I moved them around and it didn't change the noise at all.
Here's an update: I breadboarded this and found that if the center tap on the input transformer is grounded, then it is much quieter. There is still hum, but it is considerably lower. But, here's the problem -- the breadboard transformers are a cheapie pair of Edcor 10k:10k that have a center tap. But, the "real" input transformers are some nicer Cinemags that don't have a center tap ... so I am not sure what to do (I'll note that I was assured by Cinemag that no one uses the center taps, and by the designer of the circuit that they are unnecessary -- so there must be a solution). I'm also not sure what to do about the rest of the hum either, but my sense is that it must be a ground loop as that's what it sounds like - maybe the edcor is not really in the "center".
I built such an amp about four years ago. I used a pair of Hammond 125 ese in each channel. The amps were set up as if they were stereo SET amps (srpp 6sn7 driving 2a3), then I used a simple paraphase splitting arangement and wired the secondaries of the OT parallel but out of phase. They sounded pretty good, I listened to them this way for about six months, then replaced the pair of 125 ese with a single hammond 1615 pp OT, which sounded better. These amps eventually morphed into a pair of amnity 2a3.
A quick update -- it seems that the circuit is fine, and it is the user who has issues. I was using a cheap USB soundcard connected to a laptop as a source for testing, and it seems that when the laptop's power is connected to the wall, it induces this loud noise. When it runs off of batteries it is not.
How this could be makes no sense to me as this did not happen with the single ended circuit but only when two single ended circuits are connected together. Further, the laptop does not connect to safety ground so there shouldn't be a loop? And anyway, there is an input transformer which I thought was supposed to guard against this sort of thing. Any theories about what is going on would be appreciated.
How this could be makes no sense to me as this did not happen with the single ended circuit but only when two single ended circuits are connected together. Further, the laptop does not connect to safety ground so there shouldn't be a loop? And anyway, there is an input transformer which I thought was supposed to guard against this sort of thing. Any theories about what is going on would be appreciated.
dsavitsk said:
I remember having a keyboard with a metal bottom plate. In the summer I put it on my laps and it was ok until I touched the heater (central water heating). I was shocked by electricity. I checked it with voltage meter: it read 100V. Floating ground? It may be the same with laptops, at least some of them (e.g. HP old ones).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.