I am afraid this will not work. The DAC is an AC current source and will see an impedance for the I/V which is the (your example) 33 Ohm in series with the impedance of the primary winding. Your transformer will have something like few Henry up to a 100 Henry if you look at push pull transformers. so with increasing frequency, the DAC will see an ever increasing I/V impedance and the I/V voltage will clip already at low frequencies. The DAC need to see a (low) fixed value for the I/V conversion - like a fixed resistor. From here you can pick the signal for example with a transformer.
Yes, I see that issue. It is the same issue seen by a tube power amplifier with an output transformer. You would of course terminate the transformer with a resistive load which would be reflected back through the transformer to load the primary. I have to sit down and really think this through. You should be able to get more voltage gain if you let the transformer do the IV because more of the DAC’s power will be going into the transformer instead of the load resistor. The transformer will output that power on the secondary side transformed by the turns ratio and all this would depend on the proper impedances being used.
It is an idea I have had for a while but not really looked into.
I am just looking at it in terms of power delivered to the transformer. Transformers are quite efficient in terms of the power transfer from primary to secindary so what you put in comes out the other side with low loss. Right now most of the power coming out of the DAC is going to common through the load resistors. If the load is on the secondary side you have most of the power transformed to the secondary which should allow you to have a higher output voltage available at the secondary.
Looking at the AC power output of the DAC with the load resistors doing I/V - if you have 4 dac boards you have roughly .7vrms across 33 ohms on each phase. That is P=E^2/R= .7^2/33=21mw/phase or 42mw total. If you could transform that with a voltage loss of 2 DB and a secondary load of 300 ohms you would have an output voltage of 3.54 volts - 2db would give you roughly 3vrms
I have no idea if this is possible given the constraints of building transformers or if it would sound better, worse or the same as the current recommended setup.
Last edited:
You can't pass the DC through the transformer. You need the load resistors to set the DC bias at the DAC outputs.
In my original post you will see that I had specified a transformer with the correct DC resistance for the dddac. So for 4 boards it would be 33 ohms. The transformer needs to be a push pull type with a centre tap on the primary. The DCR of the transformer primary replaces the load resistors.
Happy Easter guys!
So in my constant desire to test and fry things I tested a diy tube circuit to compete against the sowters. So far so good, tubes playing lovely BUT switching back to the sowters and they are now not playing anymore, well they are but distorted like hell. Rload is fine on all 8 checking points (apos, aneg, bpos and bneg left and right against rload common on the top dac board) 33ohm biased to 40mv on a 4 deck dac.
Could it be that I somehow pumped to much DC through the transformers and they magnetized? Oo
Also tips on how to demagnetize the sowters would be great in case someone already did this with these transformers.
I checked the resistance of the secondary windings and they are fine.
Any tips on what to check or what the culprit could be? I have everything back to how it was (except new non magnetic load resistors, but as I said they measure fine, and with the tube circuit I get good sound)
Greetings Oli
So in my constant desire to test and fry things I tested a diy tube circuit to compete against the sowters. So far so good, tubes playing lovely BUT switching back to the sowters and they are now not playing anymore, well they are but distorted like hell. Rload is fine on all 8 checking points (apos, aneg, bpos and bneg left and right against rload common on the top dac board) 33ohm biased to 40mv on a 4 deck dac.
Could it be that I somehow pumped to much DC through the transformers and they magnetized? Oo
Also tips on how to demagnetize the sowters would be great in case someone already did this with these transformers.
I checked the resistance of the secondary windings and they are fine.
Any tips on what to check or what the culprit could be? I have everything back to how it was (except new non magnetic load resistors, but as I said they measure fine, and with the tube circuit I get good sound)
Greetings Oli
oo…
you do like frying things right? Well, lets see if we can remotely fix this… please start with sending a very clear picture how you connected everything.Any clue why you COULD have pushed DC through the Sowters?
ps: from Jensen:
Demagnetizing of low level transformers can generally be done with any audio generator having a continuously variable output It may take a booster of some sort to get enough level for output transformers (be sure there’s no dc offset at its output!). The idea is to drive the transformer deeply into saturation (5% THD or more) and slowly bring the level down to zero. Saturation will, of course, be easiest at a very low frequency. How much level it takes will depend on the transformer. If you’re lucky, the level required may not be hazardous to the surrounding electronics and the de-magnetizing can be accomplished without disconnecting the transformer. Start with the generator set to 20 Hz and its minimum output level, connect it to the transformer, then slowly (over a period of a few seconds) increase the level into saturation — maintain it for a few seconds — then slowly turn it back down to minimum. For the vast majority of transformers, this process will leave them in a demagnetized state.
https://www.jensen-transformers.com/wp-content/uploads/2014/08/Audio-Transformers-Chapter.pdf
Last edited:
Well perhaps I should stop testing stuff when it’s late in the evening ^^
As I wanted to be able to switch between the sowters and the tubes instantly I might have had a moment in which the Sowters were connected on pos and neg for balanced output (my normal configuration) and didn’t switch them off, so they were connected while playing through the tubes which were connected to pos and common (single ended to pre)
As I wanted to be able to switch between the sowters and the tubes instantly I might have had a moment in which the Sowters were connected on pos and neg for balanced output (my normal configuration) and didn’t switch them off, so they were connected while playing through the tubes which were connected to pos and common (single ended to pre)
Not sure if I get this right... but:
The DC between POS and NEG is zero (or very low) because the DC bias currents on the POS and NEG are equal. Therefore, connecting the transformer from POS to NEG does not do any harm; and this is actually the recommended way to connect it.
Connecting the transformer from POS to GND (or NEG to GND) is a different story. I don't remember the DC resistance of the transformer, but it might be quite low. This might result in a substantial DC current through the primary of the transformer, possibly resulting in a permanent magnetisation of the core. Is this what you think may have happened?
Do you have any means to measure the waveforms / distortion coming out of the transformer? A signal generator and a scope might be useful. Maybe you can also do this with a computer soundcard.
On another note: comparing the tubes from the SE output with the transformer balanced output is comparing apples with oranges. Can you make a tube buffer that takes a balanced input, like a long-tailed pair?
The same book also says: "Ordinary ohmmeters, especially on low-ohm ranges, can weakly magnetize the core. If an ohmmeter simply must be used, use the highest ohm range(where the test current is least)."
My first impression was that the sowters are better/more to my liking than the tubes, so I guess I won't go for balancedwith the tubes and use the tubestage in my other DAC
The transformers where never connected to ground always to POS and NEG but at times the tubestage was connected at the same time (to POS and GND) so perhaps there was some potential difference which caused a DC voltage in the sowters.
I have a shitty frequency generator and a decent oscilloscope, so I guess I will be building a demagnetisation station on my desk today.
Before doing the demagnetization I would try to see any asymmetric waveforms. Then you can check if the demagnetization helped.
Make sure the function generator output is DC free. Maybe use a (large) blocking capacitor.
Hello Italy,
It is a very nice dac. If you are not busy you should start reading the information Doede wrote on his website, read the different manuals. AND when you wanna make a change first search this thread. Better ask before you end up with a defective dac.
You can read how many problems the fifopi use has given us. There are just to many possibilities and everyone seems to have different components and different ideas as well.
Doede did a lot of a test and the safest way is to just copy one of the set ups he tested. Just a wire connected differently could change it into a non working set up.
It is a REAL task to do it in your own way and get something working like state of the art.
The changes i made in my DDDAC were only related to the power supply which is easy for most people to understand. If you end up with a supply that can deliver the right voltage at the right current the DDDAC will work, Even with very basic knowledge of electronics like me you can create something better just because you are willing to have a bigger than usual chassis. If the standard power supply uses an CRC before the regulation circuit. Depending on the number of decks you can change the R into an L. If you have a stock of chokes and just one deck it will cost you nothing. If you have 4 decks it will cost you around 100 euro. IF you are willing to spend 100 euro on a choke you might as well get a new Triad split bobbin transformer ( 20 euro max.) to change it into a LCRC supply.
Of course the DDDAC was not brought onto the market with a LCRC supply because new customers might be stopped if the total price would be 100 euro higher. There is a special upgrade supply ( with a choke) offered for a special price so to say. Same like the output options. The basic comes a 5? euro capacitor but the best option is the Sowter transformer.
SO if you start at the basic level there are several steps to make. I decided to enter at a higher level right away. Maybe less audio fun but it saves time.
The only thing i changed i going from an average 600 mH( because it needs to handle 1A with 4 decks) to one have 3000mH ( because it came available, if it would have been available 2 years ago i would bought it right away)
Greetings, Eduard
Uh nice hint thanks a lot! Checked, and yes I had only the upper half of the waveform. Everything working again, had a bad soldering spot on one Ra, weird though that the resistance measurement didn’t show this but cold solderspots can be quite magical ^^
Edit: should it be ok to connect the transformers and the tube stage at the same time or is this still a no no?
Cold solderpoints are my preferred problem because they are easy to fix
I don't know enough about your tube stage, but in general tube circuits have rather high input impedance, MUCH higher than the DC bias resistor in the DDDAC. This would not affect the DDDAC output in any way.
Back on the topic:
I found one answer to my question, not sure it is optimum but I want to share my findings with you.
I bought I/V resistors from Texas Component and fit them in my four "Tentlab" decks.
My concern about power headroom: looks OK.
Resistors stay slightly warm, something near human body temperature (didn't want to stich metallic thermocouple on them for accurate measurement)
Result:
My current setup is Foobar -> USB filter on Computer output and another one on DAC input (significantely increases quality by feeding way more accurate data)
-> Wave I/O -> Ian reclocker -> DDDAC. Revisted PSU (with kind Help of Doede) for more headroom able of 2Amps now.
For the moment listening on Headset Beyerdynamic T1+A1 as main sytem loudspeakers are in rebuild.
Texas TX2575:
Those litle things really dig into details, as a result the global balance is slightly up in the mids and the trebles.
Background noises (on lives mainly) are more understandable, voices are also gaining from this, overal restitution is cleaner.
Overall impression is as if you had used a strong solvent to remove all greasy parts in the sound.
Good on most tracks, a bit too much on others becoming slightly agressives.
In the same time the Beyerdynamic T1 is highly resolving (I call it Juge Dread ;-) ) so it should be OK on loudspeakers.
Hope this helps.
Tristan
PS: One drawback is I have to re-RIP quite some CDs
Back on the topic:
I found one answer to my question, not sure it is optimum but I want to share my findings with you.
I bought I/V resistors from Texas Component and fit them in my four "Tentlab" decks.
My concern about power headroom: looks OK.
Resistors stay slightly warm, something near human body temperature (didn't want to stich metallic thermocouple on them for accurate measurement)
Result:
My current setup is Foobar -> USB filter on Computer output and another one on DAC input (significantely increases quality by feeding way more accurate data)
-> Wave I/O -> Ian reclocker -> DDDAC. Revisted PSU (with kind Help of Doede) for more headroom able of 2Amps now.
For the moment listening on Headset Beyerdynamic T1+A1 as main sytem loudspeakers are in rebuild.
Texas TX2575:
Those litle things really dig into details, as a result the global balance is slightly up in the mids and the trebles.
Background noises (on lives mainly) are more understandable, voices are also gaining from this, overal restitution is cleaner.
Overall impression is as if you had used a strong solvent to remove all greasy parts in the sound.
Good on most tracks, a bit too much on others becoming slightly agressives.
In the same time the Beyerdynamic T1 is highly resolving (I call it Juge Dread ;-) ) so it should be OK on loudspeakers.
Hope this helps.
Tristan
PS: One drawback is I have to re-RIP quite some CDs
Thanks for the Feedback Tristan. It is always good to get some ideas around the available upgrade parts… Now waiting for someone who compare these with the Silver anti magnetic ones
PS: I use a infrared heat gun to measure temperatures in my chassis and on my boards (up to part level )
Thank you for your reply, Eduard.
I am planning to go for best bang for buck at first, I will reserve to check the LCLC psu territory in a second moment, never experienced such a psu, it will be fun and worthwhile to be implemented for sure when money will allow, will keep this in the back of my head and very likely contact you already via pm..
I shall stick to current config:
-rpi+kali reclocker
-blue motherboard
-1 deck with trans ccs+analogue tents
-15b cinemags
I have some caps to update mb and deck, some trafos and super regulators already.
After reading this thread and Doede's dddac.com site I came to following thoughts:
- i/v resistors to 270ohm susumu rg in case as I reckon I am not able to get rhopoint gg102d from this side of the pond or jump on audionote 2w tants..
-caps upgrade with oscon/silmic and various
-short i2s links
- separate left and right 12v psu for analogue, without 7810s and preregulated by 12v alwsr (andrew weekes super reg) from dual 15v toroid.
- 12v psu for the mb and deck digital side, I have a troubled salas sslv1.1 shunt psu which is in need of some cure.. and might apply a 5v alwsr super reg to the mb logic digital circuit from the start from 12v rcore trafo.
- rather standar dual 5v psu for rpi and kali from dual 12v toroid. would kali benefit from a super regulator psu? and are there worthwhile isolator solutions?
Not willing to hijack the post.. I would appreciate any hint/critique to the above.
I am planning to go for best bang for buck at first, I will reserve to check the LCLC psu territory in a second moment, never experienced such a psu, it will be fun and worthwhile to be implemented for sure when money will allow, will keep this in the back of my head and very likely contact you already via pm..
I shall stick to current config:
-rpi+kali reclocker
-blue motherboard
-1 deck with trans ccs+analogue tents
-15b cinemags
I have some caps to update mb and deck, some trafos and super regulators already.
After reading this thread and Doede's dddac.com site I came to following thoughts:
- i/v resistors to 270ohm susumu rg in case as I reckon I am not able to get rhopoint gg102d from this side of the pond or jump on audionote 2w tants..
-caps upgrade with oscon/silmic and various
-short i2s links
- separate left and right 12v psu for analogue, without 7810s and preregulated by 12v alwsr (andrew weekes super reg) from dual 15v toroid.
- 12v psu for the mb and deck digital side, I have a troubled salas sslv1.1 shunt psu which is in need of some cure.. and might apply a 5v alwsr super reg to the mb logic digital circuit from the start from 12v rcore trafo.
- rather standar dual 5v psu for rpi and kali from dual 12v toroid. would kali benefit from a super regulator psu? and are there worthwhile isolator solutions?
Not willing to hijack the post.. I would appreciate any hint/critique to the above.