There have already been some posts in the Aleph-X group buy thread on this subject, but I do not like to threadjack.
My question is, since I would like to get my Aleph-X up and running soon (as in sooner than I could concurrently build a high quality preamp), would it best to throw together a simple preamp built on perfboard with something like an LM6181 to convert from unbalanced to balanced input? Or is this not really worth the effort and expense and it is best just to run the Aleph-X with an unbalanced input until I make something fancier?
My question is, since I would like to get my Aleph-X up and running soon (as in sooner than I could concurrently build a high quality preamp), would it best to throw together a simple preamp built on perfboard with something like an LM6181 to convert from unbalanced to balanced input? Or is this not really worth the effort and expense and it is best just to run the Aleph-X with an unbalanced input until I make something fancier?
A differential (i.e. the first stage of the Aleph-X where the two IRF9610s have their Sources connected) makes a spendid phase splitter if there's enough resistance under the Sources/emitters/cathodes. The extreme endpoint of the resistance game is to use a current source (one of the few times you'll see me use one happily--it's delivering DC in that application). A current source is a fair approximation of an infinite resistance. Once you've got a current source under a differential, you can stick a signal in one side of the differential and--as if by magic--the opposite phase comes out the other side.
Okay, so there's no such thing as magic...Harry Potter notwithstanding. What happens is that the current source will literally force the current into the differential. It's easy enough to see that the driven side will use a varying amount according to whatever the signal is telling it to do. No problem, right? But what happens to the rest of the current--the unused part? It gets stuffed into the rump end of the other half of the differential. Sounds kinda rude, but they're consenting adults and what they do in the privacy of their own amplifier is their business, right? Well, that creates a signal once it gets to the load resistor for that side of the differential, even though there's no signal at the Gate.
Grey
Okay, so there's no such thing as magic...Harry Potter notwithstanding. What happens is that the current source will literally force the current into the differential. It's easy enough to see that the driven side will use a varying amount according to whatever the signal is telling it to do. No problem, right? But what happens to the rest of the current--the unused part? It gets stuffed into the rump end of the other half of the differential. Sounds kinda rude, but they're consenting adults and what they do in the privacy of their own amplifier is their business, right? Well, that creates a signal once it gets to the load resistor for that side of the differential, even though there's no signal at the Gate.
Grey
It gets stuffed into the rump end of the other half of the differential.
"Well, that creates a signal once it gets to the load resistor for that side of the differential, even though there's no signal at the Gate."
Since you have built the amp and I am afraid that I will never convince you otherwise by trying to explain the circuit, go measure the signal at the gates of the input pair with a single ended input signal. I think you are in error and I know you would not want to confuse others. Hint: Think about the current in the feedback resistors for both sides of the amp and remember no current can enter the very high impedance gates. Remember Kirchoff's Current Law: The sum the currents entering a node is equal to sum of the currents leaving a node or that all the currents in a node must equal zero. This law is as basic and as important as Ohm's law. Please don't come up with name calling and questioning my motives. I am honestly trying to help you and others understand this mode of operation. (see post above for example)
http://www.geocities.com/~supertrooper/kirch_i_n.html
"Well, that creates a signal once it gets to the load resistor for that side of the differential, even though there's no signal at the Gate."
Since you have built the amp and I am afraid that I will never convince you otherwise by trying to explain the circuit, go measure the signal at the gates of the input pair with a single ended input signal. I think you are in error and I know you would not want to confuse others. Hint: Think about the current in the feedback resistors for both sides of the amp and remember no current can enter the very high impedance gates. Remember Kirchoff's Current Law: The sum the currents entering a node is equal to sum of the currents leaving a node or that all the currents in a node must equal zero. This law is as basic and as important as Ohm's law. Please don't come up with name calling and questioning my motives. I am honestly trying to help you and others understand this mode of operation. (see post above for example)
http://www.geocities.com/~supertrooper/kirch_i_n.html
Anoher option
I picked up some opamps to do the single ended to balanced conversion a while back. I think they are DRV134. Might be DVR134.
They run on +/- 15 volt rails and recommended application notes are on the net.
I decided to go discrete and use a transformer to do this instead. But I have been told that these opamps are very nice sounding. They are fairly cheap - 4.00 USD each.
George
I picked up some opamps to do the single ended to balanced conversion a while back. I think they are DRV134. Might be DVR134.
They run on +/- 15 volt rails and recommended application notes are on the net.
I decided to go discrete and use a transformer to do this instead. But I have been told that these opamps are very nice sounding. They are fairly cheap - 4.00 USD each.
George
I also found that the Aleph sounded better (more dynamic )with the BZLS balanced ouput .Single ended out sounded a bit more constricted but I don't know if it is due to the pre sounding that way or the amp "preferring" a balanced signal.I would tend to go for the latter because subsequent experiments with single ended vs balanced connections always favour the latter.
Theoretically however if you look at the differential pair it is in itself a copy of the BZLS with a current source. So why should we add another differential pair before that? I think that perhaps the differential input pair might be upgraded so that a balanced pre is not needed with all the extra connections and distortions of an extra circuit.In my experimenting with the BZLS I found it sounds best if you bias it to the maximum possible and have no inter source resistance.The difference is very audible to my ears.
Theoretically however if you look at the differential pair it is in itself a copy of the BZLS with a current source. So why should we add another differential pair before that? I think that perhaps the differential input pair might be upgraded so that a balanced pre is not needed with all the extra connections and distortions of an extra circuit.In my experimenting with the BZLS I found it sounds best if you bias it to the maximum possible and have no inter source resistance.The difference is very audible to my ears.
Re: Anoher option
Based on this article:
http://sound.westhost.com/balance.htm
It sounded to me like a buffer stage + SSM chip would yield great results for a super low parts count. The problem with doing it in multiple op amps is that either precision resistors or resistor matching must be performed to get best results (no flame please, read the article). I considered multiple op amps, but then I agree with Grey. Once the parts count goes up significantly, a BOSOZ + CCS is likely to be the best performer for the effort required to build.
Panelhead said:
Based on this article:
http://sound.westhost.com/balance.htm
It sounded to me like a buffer stage + SSM chip would yield great results for a super low parts count. The problem with doing it in multiple op amps is that either precision resistors or resistor matching must be performed to get best results (no flame please, read the article). I considered multiple op amps, but then I agree with Grey. Once the parts count goes up significantly, a BOSOZ + CCS is likely to be the best performer for the effort required to build.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.