curious about your feedback whatever the dac chip. I like a lot the oap 1656 when some care about the parts around. Alas I am slow to refine my projects, I know I must try the opa861 pcb nice effort from Vunce with my tweaks as I like that 861 in my AYA2, story to have a more fair face to face with the TDA1541A late Taiwan or S1 either. (Burson asked to me for a try but I had to decline as not serious time enough for a benchmark and an acurate testimonial - sorry guys for that).
i had some problems for waiting parts from Mouser and shipping and had a good help from Mouser FRANCE with my few waiting parts (damn, with parts that were in stock when I triggered the purchase button but I waited 6 months) and the not so accurate work from local UPS and his customer phone service !
It is also for the guys whom asked me and I promised sooner results (sorry guys I am a snail guy but am stakanovist when it comes to the last passive parts voicing as you know) . Sorry not to be faster for @gaszto who helped me in the comming hybrid mu follower with the tubed ad1862 miro dac (and also many fellows whom helped there as a friend from Australia); I do not forgett, just I have many daily constraints that make things not easy to advance faster.
i had some problems for waiting parts from Mouser and shipping and had a good help from Mouser FRANCE with my few waiting parts (damn, with parts that were in stock when I triggered the purchase button but I waited 6 months) and the not so accurate work from local UPS and his customer phone service !
It is also for the guys whom asked me and I promised sooner results (sorry guys I am a snail guy but am stakanovist when it comes to the last passive parts voicing as you know) . Sorry not to be faster for @gaszto who helped me in the comming hybrid mu follower with the tubed ad1862 miro dac (and also many fellows whom helped there as a friend from Australia); I do not forgett, just I have many daily constraints that make things not easy to advance faster.
Last edited:
The Chinese have an AD1865 NOS module with tube output and SPDIF input, you just need to connect the transformer. 
https://www.ebay.com/itm/225093591434?hash=item34689fa18a:g:T-QAAOSwFmpi4pPi
https://www.ebay.com/itm/225093591434?hash=item34689fa18a:g:T-QAAOSwFmpi4pPi
Attachments
The first thing to take a care with simple Riv + Tube gain stage output concept is to determinate needed output voltage value.
example: if doc has +-1mA Iout that is 2mAp-p. If the Riv is 50ohm that will give 0.1Vp-p @ Riv.
And -180deg phase. Opposite phase. If the digital DATA line inverted @ Riv will be same Vout BUT 0deg in Phase.
next
IF ecc88 used in SRPP configuration gain stage vil be close to the mju, amplification factor of the tube in this case close to 33 X.
Say that is 30X. And SRPP is phase inverting circuit of -180deg.
That will give IN phase Output signal BUT Vout will be 0.1Vp-p x 30 = 3Vp-p cca.
Output resistance of the SRPP stage is about 1/S (S=transconductance), and that will be for ECC88 1 / ( 0.012 (A/V) ) = 83 ohm cca.
Output bugger is not needed.
So from the point of technical requirements for: Phase, Output Voltage and Output resistance this is a good example.
example: if doc has +-1mA Iout that is 2mAp-p. If the Riv is 50ohm that will give 0.1Vp-p @ Riv.
And -180deg phase. Opposite phase. If the digital DATA line inverted @ Riv will be same Vout BUT 0deg in Phase.
next
IF ecc88 used in SRPP configuration gain stage vil be close to the mju, amplification factor of the tube in this case close to 33 X.
Say that is 30X. And SRPP is phase inverting circuit of -180deg.
That will give IN phase Output signal BUT Vout will be 0.1Vp-p x 30 = 3Vp-p cca.
Output resistance of the SRPP stage is about 1/S (S=transconductance), and that will be for ECC88 1 / ( 0.012 (A/V) ) = 83 ohm cca.
Output bugger is not needed.
So from the point of technical requirements for: Phase, Output Voltage and Output resistance this is a good example.
My mistake the output resistance of the srpp (when same tube used) is
Ri * (Ri + Rc)/ (2*Ri+(mju+1)*Rc)
Ri = internal resistance of tube Ri=mju/S for ecc88 Ri=33/(0.0125(A/V))=2640ohm
Rc = cathode resistor say 1K in ecc88 example
.
that is
2640 x (2640+1000) / (2 x 2640 + 34 x 1000) = 9609600 / 39280 = 245 ohms
that is also good value
,
sorry
Ri * (Ri + Rc)/ (2*Ri+(mju+1)*Rc)
Ri = internal resistance of tube Ri=mju/S for ecc88 Ri=33/(0.0125(A/V))=2640ohm
Rc = cathode resistor say 1K in ecc88 example
.
that is
2640 x (2640+1000) / (2 x 2640 + 34 x 1000) = 9609600 / 39280 = 245 ohms
that is also good value
,
sorry
I don't see why it wouldn't work. It's essentially the Pass Zen I/V, with a complementary BJT voltage-follower stage added to buffer the output signal. With the values shown in the schematic, a +/- 1mA full scale DAC output current would produce a +/- 1V peak amplitude signal output.
ECC88 hybrid Mu follower aka Moglia Gyrathor, alternative to a pure CCS.
courtesy of @euro21 fellow for the TDA1541 in a thread where he helped me. Should work for the PCM63, I will never know for this last having not PCM63 or K grade either (but will not work for the AD1862, I/V resistor R is needing adaptation). D3 & D2 needs maybe to be reversed ? Not sure here ?!
Something near with a 6NP23-EV should work too before the wall is being reconstructed.
courtesy of @euro21 fellow for the TDA1541 in a thread where he helped me. Should work for the PCM63, I will never know for this last having not PCM63 or K grade either (but will not work for the AD1862, I/V resistor R is needing adaptation). D3 & D2 needs maybe to be reversed ? Not sure here ?!
Something near with a 6NP23-EV should work too before the wall is being reconstructed.
Attachments
Last edited:
I'm not a fan of SRPP itself because it has to be calculated for a specific load resistance and as such is not suitable for a universal driver. The only solution is another stage (buffer) behind the SRPP, but then there is no need for the SRPP when the classic anode amplifier stage can do it with higher gain and lower THD.
With SRPP, the bypass on the lower Rk increases the gain, reduces the output resistance but increases the distortions.
Thorsten Loesch did not use bypass on Rk, while Zoran counted with bypass.
For example in LTspice for ECC88 if Rk1,2 =1K, Riv=100R ,Idac=+/-1mA , Rload=100k (Rload=10k)
without bypass Vout=2,4Vp-p , Zout=1,5k , THD=0.036888% (0.008638%)
with bypass Vout=4,5Vp-p , Zout=300 , THD=0.146965% (0.146965%)
It can be seen from this that distortions are 4 times higher and Zout is 5 times lower if the bypass is on Rk. However, if Rload=10k(in brackets), then the THD is even 17 times higher if the bypass is over Rk. This clearly shows us the dependence of THD on Rload at SRPP.
Last edited:
Why make it complicated when you can make it simpler, in the end it comes to the same thing. This is just an idea for now. R2 and Rc should be tuned a little more to get a lower THD.
6N23P-EV is a great tube. I believe that their price will go even higher as the wall is being built, unfortunately this will happen with all the tubes very soon. I have 6N23P and 6N1P in stock , but I prefer the sound of Ei ECC88. The Ei ECC88 has nicer details and a more open sound.
