Greetings & Happy New Year!
My next project is a preamp to match my tube power amp. The issue is that the run between the preamp and the power amp is about 20 feet.
This is problematic for a single conductor high impedance cable, correct?
What is the advantage to going to a balanced line?
What losses would I have using transformers as a means to do this? What am I giving up?
I know I could do this without transformers, but the power amp is done and I do not want to change the power amp circuit, so I am resigned to look at other alternatives.
Thanks in advance!
My next project is a preamp to match my tube power amp. The issue is that the run between the preamp and the power amp is about 20 feet.
This is problematic for a single conductor high impedance cable, correct?
What is the advantage to going to a balanced line?
What losses would I have using transformers as a means to do this? What am I giving up?
I know I could do this without transformers, but the power amp is done and I do not want to change the power amp circuit, so I am resigned to look at other alternatives.
Thanks in advance!
Use 6 Meter shielded cables, with 2 signal conductors, between the preamp and the power amp. Connect the shields to ground only at the preamp end. The preamp has to have a LOW O/P impedance and good current handling capability. An IRFBC20 source follower set up for a 25 mA. drain current will do the line driver job very nicely, indeed. Yes, a miniature power amp is what the situation calls for. If some gain is needed in the line stage, a 12B4 would be excellent and DC coupling the FET to the triode is easy.
What Eli said...with the following suggestion: run the source follower in cascode; use two and deal with the variable gate capacitance. Even if D-S voltage *SEEMS* to be well outside of the high-variability zone, the cascode is worth the effort. You're going to need something that would drive a near short circuit for best results.
cheers,
Douglas
cheers,
Douglas
Thank you!
I assume that one of the inner pairs is also connected to ground at both sides of the cable, but the braided shield is only connected at the preamp end?
Now, how do I do that with tubes?
My whole point was to get all the silicon out of the loop by building the power amp and pre amp using tubes. I would imagine a cathode driven output could do that?
Last edited:
OK, I'll have to resort to Blog-whoring. 
SYclotron Audio The Heretical Preamp
Most of the stuff you need to know is in there, especially the equations for determining what you need for preamp output impedance for your situation.
SYclotron Audio The Heretical Preamp
Most of the stuff you need to know is in there, especially the equations for determining what you need for preamp output impedance for your situation.
Yes, 1 wire of the interconnect cable is "hot" and the 2nd is "cold". The shield plays no role in signal transmission.
Doing the line driver hollow state is much tougher than using FETs. You will need a high gm power pentode cathode follower. You might have to choke load, which is expensive. DC coupling between voltage amplifier and line driver will be highly problematic. Prepare to spend on PTFE film interstage coupling capacitors. IMO, FETs (darned near heaterless pentodes) are the proper choice for this job.
As I sense a determination to be SS free, a candidate for the pentode cathode follower is the 12BY7 or a similar type whose gm is higher.
I'd trust SY's judgment and experience on this. 20 feet is not that far. Single-ended will be fine. Use a good quality shielded MIC(rophone) cable with two conductors (preferably twisted pair) and a shield. Connect as described above by others.
If you already have differential out on the preamp, you might as well run differentially to the power amp.
Differential signaling will give you common-mode rejection if you implement the receiver correctly. So any noise/interference/hum injected on the cable will be rejected. Transformers are one way of converting from single-ended to differential (and back). The Jensen JT-11P is a good and reasonably priced ($70) one. The main drawback of transformers is the cost. They also have relatively high THD (around 0.01 % mid band for 0 dBu). But if you worried about 0.01 % THD you probably wouldn't be writing in the tube forum...
The trafos tend to have very high Common-Mode Rejection Ratio (CMRR). The Jensen transformer has CMRR of 130 dB. That's hard to beat.Note though, that the same job could be done by an LME49724 differential op-amp for $4 with THD below 0.00003 %. Yes. That's four zeros after the decimal point.
~Tom
Last edited:
I'd trust SY's judgment and experience on this. 20 feet is not that far. Single-ended will be fine. Use a good quality shielded MIC(rophone) cable with two conductors (preferably twisted pair) and a shield. Connect as described above by others.
If you already have differential out on the preamp, you might as well run differentially to the power amp.
Differential signaling will give you common-mode rejection if you implement the receiver correctly. So any noise/interference/hum injected on the cable will be rejected. Transformers are one way of converting from single-ended to differential (and back). The Jensen JT-11P is a good and reasonably priced ($70) one. The main drawback of transformers is the cost. They also have relatively high THD (around 0.01 % mid band for 0 dBu). But if you worried about 0.01 % THD you probably wouldn't be writing in the tube forum...
Note though, that the same job could be done by an LME49724 differential op-amp for $4 with THD below 0.00003 %. Yes. That's four zeros after the decimal point.
~Tom
Thank you!
I am still gathering information for my preamp build, so nothing is cast in stone right now.
I know Blue Jeans has audio cable at 12 pf/ft and I would guess that there are other alternatives too.
Right now I have a Dyna PS2 and too much hum.
The problem is the amp is close to the speakers, but the preamp is in a cabinet with about 20 feet of cable on it. Obviously, the mains are off of two different sockets, so, ground loops prevail.
The amp also has a 10K input impedance, which I guess I should change to a higher impedance?
I need a way to drive across a 20' line and I need a way to kill the ground loop hum I get. Those are my goals.
Well, some quick calculation will show you that with even prosaic coax, a source impedance of a few hundred ohms should be satisfactory- see the Heretical article.
Input transformers at the power amp end will galvanically isolate things. A more elegant approach is to fix the grounding in your pre and power amps so that the ground loops are interrupted. That's usually done by lifting the signal/PS grounds from the chassis/safety ground and using a groundbreaker (e.g., two antiparallel diodes) between the signal ground point and the chassis.
The Dyna is probably not worth the effort to do that to.
However, the power amp has a star ground to the chassis. Both the power supply and the audio circuit tie into that star. The star is bolted to the aluminum chassis along with the safety ground from the line.
If I read your post correctly, I should lift that star grounding post off of the chassis and electrically connect it back to the chassis (and safety ground from the mains) via two diodes in series with the anodes facing each other.
Is that correct?
What type of diodes do I need?
However, the power amp has a star ground to the chassis. Both the power supply and the audio circuit tie into that star. The star is bolted to the aluminum chassis along with the safety ground from the line.
If I read your post correctly, I should lift that star grounding post off of the chassis and electrically connect it back to the chassis (and safety ground from the mains) via two diodes in series with the anodes facing each other.
Is that correct?
What type of diodes do I need?
Last edited:
Just to be sure:
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.