Here's a problem that I've been puzzling over for a while. I sure would appreciate some other thoughts, because I'm not able to solve it.
There are a number of ways to go from unbalanced to balanced signals. One of them is a simple split load (concertina) phase splitter. If both outputs of this topology are loaded properly it can give a very nice balanced output when the signal is taken between the outputs. Here's a picture just for reference, even though I know you guys don't need it.
But, what I'm trying to figure out is this. Even though the gain of each leg is less than one (for most triodes that we use, let's say .9), the "effective" gain when you take the balanced output is twice this number (say 1.8). That is, with 1V unbalanced coming in there will be 1.8V balanced coming out measured across the outputs.
What I'm trying to figure out is this: is there a way to have unity gain from unbalanced to balanced. 1V unbalanced in gives 1V balanced out. Is there a way to do this other than the simple technique of using a triode with low enough mu so that each leg has only a gain of .5? In other words, can this topology be made to look like a transformer with a 1:1 ratio?
Thanks for the help.
There are a number of ways to go from unbalanced to balanced signals. One of them is a simple split load (concertina) phase splitter. If both outputs of this topology are loaded properly it can give a very nice balanced output when the signal is taken between the outputs. Here's a picture just for reference, even though I know you guys don't need it.
But, what I'm trying to figure out is this. Even though the gain of each leg is less than one (for most triodes that we use, let's say .9), the "effective" gain when you take the balanced output is twice this number (say 1.8). That is, with 1V unbalanced coming in there will be 1.8V balanced coming out measured across the outputs.
What I'm trying to figure out is this: is there a way to have unity gain from unbalanced to balanced. 1V unbalanced in gives 1V balanced out. Is there a way to do this other than the simple technique of using a triode with low enough mu so that each leg has only a gain of .5? In other words, can this topology be made to look like a transformer with a 1:1 ratio?
Thanks for the help.
Attachments
You dont need this trick to achieve equal output impedances in a split load phase inverter, as long as both outputs are equally loaded also the output impedances are equal, this has been debated and described many times in this forum.
However you are correct in that splitting anode and cathode resistors is one way of achieving lower gain, but I want to know why Runeight want to achieve this, if needed i would probably put a resistive splitter on the input.
Regards Hans
A fair question Hans. Perhaps I should describe the problem first. 
I have a pair of monoblocks that are quite sensitive. They only need 1v peak to reach full power. But, their inputs are balanced.
So, I don't need a preamp, just an attenuator for most reasonable sources. I know that there are CD players with balanced outputs and I can go there, but they are generally more expensive than average CD player. I don't know about phono sources yet because I have not looked into their availability.
Assumming then that the sources are single-ended, I need to translate from unbalanced to balanced without adding any gain.
I've looked at transformers and can easily go this way. But, I'm not in a terrible hurry so I'm just looking at other ways to get from unbalanced sources to balanced amps.
of course, the resistive splitter on the input is just the volume control. but, for many sources, i'll barely be able to crack the volume if the preamp itself has a gain of 1.8 or so. i could put another divider at the input and probably wouldn't suffer too much from doing it, perhaps a little less s/n. but at these levels not necessarily an issue.
And since the "preamp" doesn't need any gain, my question.
tschrama, i can indeed use voltage dividers in the loads. there are some problems that occur in doing this, but this is one possible option.
I have a pair of monoblocks that are quite sensitive. They only need 1v peak to reach full power. But, their inputs are balanced.
So, I don't need a preamp, just an attenuator for most reasonable sources. I know that there are CD players with balanced outputs and I can go there, but they are generally more expensive than average CD player. I don't know about phono sources yet because I have not looked into their availability.
Assumming then that the sources are single-ended, I need to translate from unbalanced to balanced without adding any gain.
I've looked at transformers and can easily go this way. But, I'm not in a terrible hurry so I'm just looking at other ways to get from unbalanced sources to balanced amps.
of course, the resistive splitter on the input is just the volume control. but, for many sources, i'll barely be able to crack the volume if the preamp itself has a gain of 1.8 or so. i could put another divider at the input and probably wouldn't suffer too much from doing it, perhaps a little less s/n. but at these levels not necessarily an issue.
And since the "preamp" doesn't need any gain, my question.
tschrama, i can indeed use voltage dividers in the loads. there are some problems that occur in doing this, but this is one possible option.
Sowter TVC
Hello ,
I'm currently using a Sowter TVC . These can be configured to provide an unbalanced to balanced conversion and can be driven directly by a CD player . It sounds very good in comparison to resistive attenuators too
cheers
316a
runeight said:A fair question Hans. Perhaps I should describe the problem first.
I have a pair of monoblocks that are quite sensitive. They only need 1v peak to reach full power. But, their inputs are balanced.
So, I don't need a preamp, just an attenuator for most reasonable sources. I know that there are CD players with balanced outputs and I can go there, but they are generally more expensive than average CD player. I don't know about phono sources yet because I have not looked into their availability.
Assumming then that the sources are single-ended, I need to translate from unbalanced to balanced without adding any gain.
I've looked at transformers and can easily go this way. But, I'm not in a terrible hurry so I'm just looking at other ways to get from unbalanced sources to balanced amps.
of course, the resistive splitter on the input is just the volume control. but, for many sources, i'll barely be able to crack the volume if the preamp itself has a gain of 1.8 or so. i could put another divider at the input and probably wouldn't suffer too much from doing it, perhaps a little less s/n. but at these levels not necessarily an issue.
And since the "preamp" doesn't need any gain, my question.
tschrama, i can indeed use voltage dividers in the loads. there are some problems that occur in doing this, but this is one possible option.
Hello ,
I'm currently using a Sowter TVC . These can be configured to provide an unbalanced to balanced conversion and can be driven directly by a CD player . It sounds very good in comparison to resistive attenuators too
cheers
316a
Am I missing something?
If I understand you correctly, feeding each of your monobloc amps from a balanced input of 1v RMS each side will cause them to give full output. Feeding them from a shared single-ended 1v RMS input via a concertina splitter should give ~0.9v RMS per side. I don't see why this is a problem.
If I understand you correctly, feeding each of your monobloc amps from a balanced input of 1v RMS each side will cause them to give full output. Feeding them from a shared single-ended 1v RMS input via a concertina splitter should give ~0.9v RMS per side. I don't see why this is a problem.
Yes, I am sure but unfortunately many text books and articles in magazines and on the net get it wrong. The big mistake almost everyone does is to ignore the fact that a phase splitter is loaded on both outputs, if you assume that one output is unloaded you get the result that it behaves as an ordinary amp stage with much lower impedance on cathode than on anode.
However in real life application as long as the output tubes doesn't stray into A2 or AB2, (where grid current is drawn and the impedance varies with the envelope) the output impedance from anode and cathode side is equal and not only that, it is lower than from an ordinary cathode grounded stage with the same circuit values.
Even Morgan Jones got it wrong in earlier revisions of his book but now he has corrected this and explains in an easy way why the impedeances are equal. An even better explanation can be found here: http://www.aikenamps.com/cathodyne.pdf
I have built and simulated several phase splitters and confirmed that the described theory is correct.
It is very unfortunate that the misunderstanding of output impedance of this phase splitter is so well spread, I think it would have been used even more if designers understood it better.
If the output tubes go into A2 or AB2 the output impedances changes, but that is not so different from any other phase splitter circuit, A2 or AB2 amplifiers should have separate driver stages anyway.
Regards Hans
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.