John,
The board will have a + & - terminal from the transformer for the lower voltage, and three terminals for the higher voltage to power the opamp output stage [probably marked 15+ / 0 /15- volts]. it is from the - in the case o9f my board - terminal marked 0 of these three I take two wires to the terminal on the A-20 transformers which is marked with an 'earth' symbol. Without this there was a little noise but with these connections not only did the noise disappear but the treble quality was much improved.
For the a-20 the earth terminal almost certainly connects to the core.
I do NOT use USB but a CD transport. Can you lay your hands on a CD transport / player with a RCA digital out? It is possible that the PC/Mac contains the root of the problem.
APOLOGIES: Posted without reading Bill and Rossi!!
BC
Last edited:
Yes, I agree with that statement.
In most cases I think we would agree that the shield would be tied to the circuit ground and not earth. I'm hoping some of the transformer guys will chime in on that one.
The metal casing can be tied to circuit ground also. If the transformer is small enough that it can be mounted on the PCB, thats a good idea.
If I read brianco's message correctly, he is doing that.
Physically large output transformers may need to be mounted on the chassis and that will have a lot of capacitance to the chassis. I wouldn't worry about it too much to have a big 600 ohm output transformer mounted to earth ground.
here are some more links useful to the discussion. These are found on the jensen web:
Hum & Buzz in Unbalanced Interconnect Systems
http://www.jensen-transformers.com/an/an004.pdf
Additional RFI Protection for Line Input Circuits
http://www.jensen-transformers.com/as/as072.pdf
Proper Grounding Inside Equipment Avoids Ground Noise Coupling
http://www.jensen-transformers.com/as/as085.pdf
rossl:
Nice picture. Given you have galvanic isolation between each component, I would suggest the best way to minimize noise injection is to bond with zero resistance, not 10 ohms. Not unlike grounding/bonding in power systems; the higher in resistance you bond with, the greater your risk of transient overvoltages and coupling through stray capacitance. By connecting with zero resistance, you effectively short out the strays. Allowing resistance in the circuit permits the build up of voltage (static or inductive) which translates into current flow across the stray impedances.
RF is of course a whole 'nother ballgame. You might find improved performance with ferrite beads or a ceramic cap from PS negative to 'earth'. There impedance is your friend to limit antenna action.
There are some good app notes from Jensen Transformers on this, if I recall. Just my $0.02
Nice picture. Given you have galvanic isolation between each component, I would suggest the best way to minimize noise injection is to bond with zero resistance, not 10 ohms. Not unlike grounding/bonding in power systems; the higher in resistance you bond with, the greater your risk of transient overvoltages and coupling through stray capacitance. By connecting with zero resistance, you effectively short out the strays. Allowing resistance in the circuit permits the build up of voltage (static or inductive) which translates into current flow across the stray impedances.
RF is of course a whole 'nother ballgame. You might find improved performance with ferrite beads or a ceramic cap from PS negative to 'earth'. There impedance is your friend to limit antenna action.
There are some good app notes from Jensen Transformers on this, if I recall. Just my $0.02
Hi.
In addition to connecting a wire from the 0V connection on the DAC board, to the earth terminals on the A20s (marked CT on the transformer wiring guides on the sides), do you also connect from the earth terminal/CT to pins 3&4, and pins 9 & 10??
Many thanks,
- John
P.S. I DO have this correct don't I? The CT mentioned on the transformer sides IS the earth terminal, or have I got this all wrong, and it's the connected 3&4, and 9&10 pins which are the CT, and the earth terminal is, well the earth terminal?!? Thanks
Last edited:
I'm very interested in trying out the schematic in the second diagram as I have the upsampling DAC mentioned in another thread here that I have fitted UTC A-20 transformers to. I also have the 2x100 Sure electronics tripath board that has become popular over in the Class D forum. Both of my sources have variable output(Oppo and Squeezebox). So I can bypass the 2.2 uf cap I have put on the input of the T-amp and go directly in with the UTC? Any idea where the 2.5 volt connection would be on this board?
I hope I'm not hijacking this but it seemed related to the topic.
Hi vaughn,
I don't have the Sure electronic board just the Lepai Tripath TA2020 based amp. Is your amp a TA2020? If so, then pin 14 is called biascap & has the 2.5V on it - it's connected to ground via a cap on the Lepai - probably the same on your board. You need to trace this and in my case (I haven't done this yet - too many different things to test, fix,etc) I would lift this cap (smd) and attach a lead to the pad on the pin 14 side. This lead woul now be at 2.5V & connected to one side of the secondary of the transformer as per Pano diagram.
Hope this helps & if not correct, I'm sure someone will jump in.
Do post your impressions here when you have tested it!
How are you finding the UTCs on DAC o/p?
I don't have the Sure electronic board just the Lepai Tripath TA2020 based amp. Is your amp a TA2020? If so, then pin 14 is called biascap & has the 2.5V on it - it's connected to ground via a cap on the Lepai - probably the same on your board. You need to trace this and in my case (I haven't done this yet - too many different things to test, fix,etc) I would lift this cap (smd) and attach a lead to the pad on the pin 14 side. This lead woul now be at 2.5V & connected to one side of the secondary of the transformer as per Pano diagram.
Hope this helps & if not correct, I'm sure someone will jump in.
Do post your impressions here when you have tested it!
How are you finding the UTCs on DAC o/p?
Thanks for the reply, I have this amp...
http://www.diyaudio.com/forums/showthread.php?t=143669
It uses the TP2050 chip so I don't think it is going to have the same pinout as the 2020. I suppose I could find the corresponding pin in the data sheet for the
2050.
The UTC trannies have take this DAC to another level. It was (unmodded) the best sounding source I have ever had. With the transformer output it has suddenly become the strongest link in my audio chain. I am very happy with it but wish I had the option of switching between digital sources(Oppo DVD and Squeezebox), both of which have digital volume control and would allow me to run the DAC straight into the amp. I am now lusting after the Behringer DAC, of course, which would allow me to do this.
http://www.diyaudio.com/forums/showthread.php?t=143669
It uses the TP2050 chip so I don't think it is going to have the same pinout as the 2020. I suppose I could find the corresponding pin in the data sheet for the
2050.
The UTC trannies have take this DAC to another level. It was (unmodded) the best sounding source I have ever had. With the transformer output it has suddenly become the strongest link in my audio chain. I am very happy with it but wish I had the option of switching between digital sources(Oppo DVD and Squeezebox), both of which have digital volume control and would allow me to run the DAC straight into the amp. I am now lusting after the Behringer DAC, of course, which would allow me to do this.
I am planning transformer volume controls, with a gap, and intend to use them as both input and output to the Berringer DCX. A friend has pointed out that the Berringer has DAC's with built in op amps, so he suggested I replace those devices with new ones. With a gap in the transformer core and 80% nickle core, I expect these to be a really fine addition. It will require another box on top I suspect.
Bud
Bud
I've been bold 
- I posted progress on another thread so I wanted to update this one & ask a question or two.
I hooked up some Lundahl LL1517 transformers but in order to compare them properly to the Slagle transformer, I used before, I need to have both operating at their theoretical best.
So here's some factors I need to reconcile in order to get the best out of each transformer:
- the PCM1793 DAC I'm using, wants to drive into a minimum of 1.8K ohms
- the Slagle transformer will have minimum thd @ 20Hz if the primaries are loaded with 8K - is that 4K on each leg in series or 8K across the primary? Is it OK to present a 8K load to the DAC or will it throw away some of the signal?
- the LL1517 works best with a minus 18 ohm impedance load - can this only be achieved with op-amp mixed feedback circuit as suggested in the datasheet?
The progress I made is in connecting the DAC output transformers to the inputs of a TA2020 Tripath amplifier without caps. It extends both LF & HF, particularly noticeable at HF, removing the smear of the caps I was using.
Using the Slagle trafo there is a HF hiss now in evidence that I need to track down which is not so evident in the Lundahls. Is this the HF extension of the Slagle's showing up some HF issues I have in the system?
Could this hiss be DC hash coming from the laptop feeding the USB DAC. I tried not having the laptop grounded but that changed nothing. I tried disconnecting the USB ground to the laptop, just connecting to earth but my DAC doesn't work in that configuration.
Does anybody know from where I can buy one of those Analog Devices ADUM4160 USB isolator chips? I know they will not deal with 24/192 speed but I figured worth a try!
- I posted progress on another thread so I wanted to update this one & ask a question or two.I hooked up some Lundahl LL1517 transformers but in order to compare them properly to the Slagle transformer, I used before, I need to have both operating at their theoretical best.
So here's some factors I need to reconcile in order to get the best out of each transformer:
- the PCM1793 DAC I'm using, wants to drive into a minimum of 1.8K ohms
- the Slagle transformer will have minimum thd @ 20Hz if the primaries are loaded with 8K - is that 4K on each leg in series or 8K across the primary? Is it OK to present a 8K load to the DAC or will it throw away some of the signal?
- the LL1517 works best with a minus 18 ohm impedance load - can this only be achieved with op-amp mixed feedback circuit as suggested in the datasheet?
The progress I made is in connecting the DAC output transformers to the inputs of a TA2020 Tripath amplifier without caps. It extends both LF & HF, particularly noticeable at HF, removing the smear of the caps I was using.
Using the Slagle trafo there is a HF hiss now in evidence that I need to track down which is not so evident in the Lundahls. Is this the HF extension of the Slagle's showing up some HF issues I have in the system?
Could this hiss be DC hash coming from the laptop feeding the USB DAC. I tried not having the laptop grounded but that changed nothing. I tried disconnecting the USB ground to the laptop, just connecting to earth but my DAC doesn't work in that configuration.
Does anybody know from where I can buy one of those Analog Devices ADUM4160 USB isolator chips? I know they will not deal with 24/192 speed but I figured worth a try!
Well, best to ask Kevin Karter about the Lundahl devices. Per Lundahl runs a very correct engineering firm, so I would trust their comments upon best usage.
And, of course, Dave will be the person to comment on your usage and findings when using his transformer. Though, you may be contributing to a tendency to peak high frequencies when its primaries are this badly mismatched to their load. The 8 K would typically be a full primary specification, for a push pull or single ended OPT.
You might PM Panomaniac about your questions also. He has considerable experience with transformer loading DAC chips. You might also find his comments about my transformers used in this fashion entertaining.
And, finally, there is not a sand amp in the world that will not be improved in quality of reproduced music, by the addition of a good OPT. This has little to do with distortion specs and the usual complaint in that direction. Sand just loves iron. Ask Pano about Gary Pimms SS PP all transformer coupled amplifier, and what it is capable of.
Bud
And, of course, Dave will be the person to comment on your usage and findings when using his transformer. Though, you may be contributing to a tendency to peak high frequencies when its primaries are this badly mismatched to their load. The 8 K would typically be a full primary specification, for a push pull or single ended OPT.
You might PM Panomaniac about your questions also. He has considerable experience with transformer loading DAC chips. You might also find his comments about my transformers used in this fashion entertaining.
And, finally, there is not a sand amp in the world that will not be improved in quality of reproduced music, by the addition of a good OPT. This has little to do with distortion specs and the usual complaint in that direction. Sand just loves iron. Ask Pano about Gary Pimms SS PP all transformer coupled amplifier, and what it is capable of.
Bud
Thanks Bud,
I talked to you before about the SS Tabor & have all the CCS boards built & ready to go. I need a hefty heatsink & more time (and possibly experience) to complete it but based on your comments on the audition in Gary's house and the version 1 you heard there, I was enthused!
My Slagle trafos are ear-marked for a John Swenson BDT pre-amp that was going to feed into the SS tabor.
I will make contact with all whom you suggest, thanks.
I talked to you before about the SS Tabor & have all the CCS boards built & ready to go. I need a hefty heatsink & more time (and possibly experience) to complete it but based on your comments on the audition in Gary's house and the version 1 you heard there, I was enthused!
My Slagle trafos are ear-marked for a John Swenson BDT pre-amp that was going to feed into the SS tabor.
I will make contact with all whom you suggest, thanks.
Hi Mike (Panomaniac),
Thanks for all the contribution on the thread. I'll ask my question here so others might benefit from the answer.
Your transformer schema shows a 2k R across the primary of the trafo which swamps the DCR & flattens the kins in the impedance, correct? This will look like a flattish 1K load to the DAC if using a 1:1 600R transformer. Mine is 400R so how do I calculate the apparent load to the DAC - it's not just Rs in parallel?
Have you used Lundahl LL1517s before I fire a Q off to KevinC at K&K audio about them? The negative impedance issue is bugging me.
Thanks for all the contribution on the thread. I'll ask my question here so others might benefit from the answer.
Your transformer schema shows a 2k R across the primary of the trafo which swamps the DCR & flattens the kins in the impedance, correct? This will look like a flattish 1K load to the DAC if using a 1:1 600R transformer. Mine is 400R so how do I calculate the apparent load to the DAC - it's not just Rs in parallel?
Have you used Lundahl LL1517s before I fire a Q off to KevinC at K&K audio about them? The negative impedance issue is bugging me.
What is "SS PP all transformer coupled amplifier"?
I'am curious about it, can i have more info?
Ciao
Andrea
I had no idea!
...so if I can control the volume level out of my Squeezebox or Oppo when I am using their digital outputs I assume the volume control is "digital". You say that this is actually attenuating the sound quality! I'm glad you mentioned this, of course now it complicates my life
Interesting that this is not widely known, but, when you digitally attenuate a signal, you throw away bits. No other way to accomplish this. Buy a TVC for your upstream master volume control, see if you can run your preamp wide open and remove it's volume control.
Gary Pimm's push pull solid state all transformer coupled amp is one of the two best amplifiers I have ever heard, and I was using EnABL'd speakers at the time, a three way full range system.
Bud
Gary Pimm's push pull solid state all transformer coupled amp is one of the two best amplifiers I have ever heard, and I was using EnABL'd speakers at the time, a three way full range system.
Bud
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.