Back in the early 80s we used to run RS232 terminals at a much longer distance (hundreds of meters) than supposedly was possible. We used good quality shielded cable and attached the shield to chassis ground at one end of the cable only, thus avoiding loops.
Using that same method (for single ended use) you would use the 2 insulated wires in standard microphone/balanced cable for signal and gnd and would connect the shield at one side only to chassis gnd. After my recent experiences with radiated noise in a test setup of a preamp, I will be trying this method myself. YMMV 🙂
This it it twisted shielded wires I assume?
Yes, it should be very good. The shield protect against RF and the twisting against magnetic induced noise in the cable.
I think this is how most HighEnd RCA cables are made but you could make the same much cheaper.
I have been lazy and I need some very good RCA connectors anyway (twist lock preferred). So I say......100 USD is ok for at cable if connectors are top quality.
But interesting to "DIY".......
About cables:
https://ing.dk/holdning/audiohomoeopater-og-kabeltosser
The Danish guy "Poul-Henning Kamp" is "burdened by concrete knowledge" and the thread is quite fun to read and his arguments.......but you need to learn Danish or use google translate 🙂
Yes, we used twisted pair for those terminals in those days.
Ideally I would use twisted pairs, but I recently saw and bought 5 meters of cheap 30AWG UL2547 shielded (but not twisted) cable at Aliexpress for testing. At under US$4 for 5m I could not resist to try it out for that preamp I am testing 🙂
Ideally I would use twisted pairs, but I recently saw and bought 5 meters of cheap 30AWG UL2547 shielded (but not twisted) cable at Aliexpress for testing. At under US$4 for 5m I could not resist to try it out for that preamp I am testing 🙂
mains transformers are for Sissies
I'm powering my preamps with sheer will
I'm powering my preamps with sheer will
|
would one good gemini PS be able to feed 4 buffer cells instead of 2 as done in Iron Pre?
Meaning 2 X SE stereo pre (each one with independent capmultiplier) with only one gemini PS with doubled current programmed ?
Meaning 2 X SE stereo pre (each one with independent capmultiplier) with only one gemini PS with doubled current programmed ?
because I would like to ad a second buffer cell per channel without edcors to make an active crossover for the tweeters
going biamp
going biamp
Last edited:
Why 36 and not 2x18?
TTSA0030 - Transformer AUDIO TSA30VA - voltage to 50VOutput voltage: 36 V; Mounting pad: Filled epoxy resin #07
This looks ok.
Gae.
Finally got my boards and I have a question about layout. What's more important, keeping the boards away from the transformers, or placing the volume pot closer to the back of the case and therefore having shorter wires? Hope that makes sense.
keep the boards away from the transformers and place the volume pot closer to the back of the case
The Iq of the JFET buffer is close to Idss of the pair (just a bit below caused by the 20R pot so Vgs is not zero (in idle))?
Is it an advantage to have JFET pairs with high Idss to maybe have better "drive" of the Cinemag?
Is it an advantage to have JFET pairs with high Idss to maybe have better "drive" of the Cinemag?
just look at graphs showing xconductance in function of Iq, for all 3 ranges of Idss
simple as that
using duals and triplets, so paralleling more to achieve greater xconductance .......... that's worth thinking
simple as that
using duals and triplets, so paralleling more to achieve greater xconductance .......... that's worth thinking
It is my understanding that 2SK370 is the same "chip" as 2SK170 and the difference in current (200 vs 400 mA) is because the larger case of 2SK170 gives better cooling? ......so I can use 2SK170 data to look at?
How tight Idss match is necessary to be able to parallel? -+0.5 mA?
How tight Idss match is necessary to be able to parallel? -+0.5 mA?
It would be quite easy to parallel as one set of JFETs can be soldered at the backside of PCB.
Do you think the 2SK/2SJ octets sold by DiyStore are close enough for paralleling?
Then I would need a "double" octet for a BAL pre.
I am only allowed to buy one octet pr. month 🙂
I would be good to get 2 x octets that are relative close.......for symmetry reasons.
Do you think the 2SK/2SJ octets sold by DiyStore are close enough for paralleling?
Then I would need a "double" octet for a BAL pre.
I am only allowed to buy one octet pr. month 🙂
I would be good to get 2 x octets that are relative close.......for symmetry reasons.
paralleling, keep them in 0.1mA bracket
you can easily solder 2 JFets in pads of one, with some care, no need for piggybacking
200 vs. 400 - you mean mW, not mA, so yes - case is difference
you can easily solder 2 JFets in pads of one, with some care, no need for piggybacking
200 vs. 400 - you mean mW, not mA, so yes - case is difference
Yes, I meant mW....
So you think there is physical space enough on PCB on front side to parallel......even if 2SK170 / 2SJ74 are used......
I wonder if DiyStore 2SK/2SJ octets are 0.1 mA matched......I don't think they specify it.
Next is to be able to buy two octets that are close to each other.
So you think there is physical space enough on PCB on front side to parallel......even if 2SK170 / 2SJ74 are used......
I wonder if DiyStore 2SK/2SJ octets are 0.1 mA matched......I don't think they specify it.
Next is to be able to buy two octets that are close to each other.