I dont think i have any lying around...i will check and see.
Im getting 44.1khz before the 3.3k resistor on the FS line and 0.02/0.03khz just after it.
Thanks for the help guys...i appreciate it
Alon
Thanks Oliver
that would be a great help. I think im gonna take a break now...my head is hurting!!Will wait till you try it out Oliver.
Thanks
Alon
Hi Oliver,
How did you select the dip switch?
I get varying frequencies from the dout now...the way i have set mine is like this.
10+
(AIFCON0) on middle or 0
(AIFCON1) on middle or 0
(TXS/RC) on middle or 0
(AIF/MS) to the right or + (before i made the mistake of going to the left or 1)
Try that and see if you get some frequencies on dout.
Alon
How did you select the dip switch?
I get varying frequencies from the dout now...the way i have set mine is like this.
10+
(AIFCON0) on middle or 0
(AIFCON1) on middle or 0
(TXS/RC) on middle or 0
(AIF/MS) to the right or + (before i made the mistake of going to the left or 1)
Try that and see if you get some frequencies on dout.
Alon
Last edited:
Hmmm strange...i just checked mine again...and with a signal playing the frequencies change on my Dout...as soon as i stop the signal...the frequencies go back to 0.01 or there abouts.
Hi guys, I know I've been absent for a long while. I've not had time to play around with my dac much, but over the last few days I've had a proper go at it. I have a couple of issues and could do with a little help if anyone can.
I've installed more psu's and tx's and every thing is sounding great. BUT:
The DEM reclocking is not working (so far as I can tell). It doesn't matter what frequency I run the DEM reclocking on it sounds exactly the same. I thought this was strange so I removed it altogether and again, no change. However, if I remove the -15v feed from the 2k2 resistors I get distortion in the sound, with or without a clock signal feeding the DEM.
Can anyone think of why this would be?
Cheers, Lee.
I've installed more psu's and tx's and every thing is sounding great. BUT:
The DEM reclocking is not working (so far as I can tell). It doesn't matter what frequency I run the DEM reclocking on it sounds exactly the same. I thought this was strange so I removed it altogether and again, no change. However, if I remove the -15v feed from the 2k2 resistors I get distortion in the sound, with or without a clock signal feeding the DEM.
Can anyone think of why this would be?
Cheers, Lee.
Guys, I have just powered up the psu pcb's. Weird thing is that the bleeder does not seem to work. I have Kiwames 270K, but there is (almost) no bleeding so the caps banks stay full for hours/days. Is that how its supposed to be?
Also the Kiwames seem to measure unlimited resistance (at least larger than my dmm can measure), but since all 4 have that, I expect it to be a measurement error.
Also the Kiwames seem to measure unlimited resistance (at least larger than my dmm can measure), but since all 4 have that, I expect it to be a measurement error.
Hi dvb-projekt,
I assume it's this one,
The DEM clock must be tested with low level signals / recordings:
Download Sheffield Lab Test Series, The - My Disk (The Sheffield/Autosound 2000 Test Disc) 2000 APE
Use track 37 (Music recorded with transfer level of -40 dB). Volume control probably has to be set almost to maximum, and there should be lowest possible ambient noise in the listening room. You should now be able to hear the music recording while listening close to one of the speakers.
The music can be a bit noisy (depending on audio set SN ratio), but with correct working DEM circuit the music should not be distorted. If the music distorts, the DEM circuit is not working correctly. If the distortion is already audible at normal sound levels, the DEM clock has probably stopped and the 6 MSBs are no longer accurate.
The balanced DC-coupled DEM circuit works with various DEM clock frequencies. Systems that offer high enough resolution can differentiate between the subtle differences at different DEM clock rates. In general, the higher DEM clock rates (up to 2.8224 MHz) offer slightly smoother sound on a high resolution system.
If no audible difference between various DEM clock rates, audio set resolution is too low. This however doesn't mean that the DEM clock isn't working.
When no balanced clock signal is applied (2K2 resistors still in place) the DEM oscillator oscillates near maximum frequency (around 7 MHz). As long as this oscillation remains stable, the sound will be ok. When both 2K2 resistors are removed, oscillation stops or becomes intermittent (depends on stray capacitance between pins 16 and 17). In this case distortion occurs.
I assume it's this one,
The DEM clock must be tested with low level signals / recordings:
Download Sheffield Lab Test Series, The - My Disk (The Sheffield/Autosound 2000 Test Disc) 2000 APE
Use track 37 (Music recorded with transfer level of -40 dB). Volume control probably has to be set almost to maximum, and there should be lowest possible ambient noise in the listening room. You should now be able to hear the music recording while listening close to one of the speakers.
The music can be a bit noisy (depending on audio set SN ratio), but with correct working DEM circuit the music should not be distorted. If the music distorts, the DEM circuit is not working correctly. If the distortion is already audible at normal sound levels, the DEM clock has probably stopped and the 6 MSBs are no longer accurate.
The balanced DC-coupled DEM circuit works with various DEM clock frequencies. Systems that offer high enough resolution can differentiate between the subtle differences at different DEM clock rates. In general, the higher DEM clock rates (up to 2.8224 MHz) offer slightly smoother sound on a high resolution system.
If no audible difference between various DEM clock rates, audio set resolution is too low. This however doesn't mean that the DEM clock isn't working.
When no balanced clock signal is applied (2K2 resistors still in place) the DEM oscillator oscillates near maximum frequency (around 7 MHz). As long as this oscillation remains stable, the sound will be ok. When both 2K2 resistors are removed, oscillation stops or becomes intermittent (depends on stray capacitance between pins 16 and 17). In this case distortion occurs.
Did you measure it in the complete circuit or stand alone?
I made a measurement of the -25V line in the circuit and after power off the voltage drops
within one minute under -3V. I think that´s ok. If you want to have a faster voltage drop,
you could surely lower the bleeder value.
Thank´s John for the DEM-Reclock answer!
Last edited:
I measured it stand alone.
In theory:
Total capacitance per bank is 15 * 1800 uF = 27.000 uF
C = F * V = 27.000 * 29 = 783.000 (on the 25V bank)
A = V/R = 29 / 270.000 = 0.107 mA
S = C / A = 783 / 0.107 = 7317 seconds until discharged = 2 hours (not linear)
In practice it takes much longer, as this morning there was still voltage (2V or so) left on the rails...
You are right that in the circuit it is of course significantly faster.. Just a bit annoying when debugging/testing that I need to put in an additional bleeder to discharge.
ah well, not the most important I guess.
In theory:
Total capacitance per bank is 15 * 1800 uF = 27.000 uF
C = F * V = 27.000 * 29 = 783.000 (on the 25V bank)
A = V/R = 29 / 270.000 = 0.107 mA
S = C / A = 783 / 0.107 = 7317 seconds until discharged = 2 hours (not linear)
In practice it takes much longer, as this morning there was still voltage (2V or so) left on the rails...
You are right that in the circuit it is of course significantly faster.. Just a bit annoying when debugging/testing that I need to put in an additional bleeder to discharge.
ah well, not the most important I guess.
Thanks for the reply John.
The resolution of the kit up here isn't that great to tell the truth (it's just my office system). I was just concerned after reading so many positive things about how great the DEM clocking was that I couldn't discern a difference between no reclock, /2 and /4.
I thought though that the DEM reclock would work without the 2k2 resistors. With no 2k2 resistors connected I get distortion no matter whether I run /2 or /4 into it. Should I try even lower frequencies?
Thanks, Lee.
The resolution of the kit up here isn't that great to tell the truth (it's just my office system). I was just concerned after reading so many positive things about how great the DEM clocking was that I couldn't discern a difference between no reclock, /2 and /4.
I thought though that the DEM reclock would work without the 2k2 resistors. With no 2k2 resistors connected I get distortion no matter whether I run /2 or /4 into it. Should I try even lower frequencies?
Thanks, Lee.
This is a DC-coupled balanced DEM injector, The 2K2 resistors form a voltage divider together with the 12K resistors. This divider also ensures correct bias voltage (around -10 ... -11V) on pins 16 and 17.
When the 2K2 resistors are disconnected, this bias voltage is gone, and both 12K resistors pull the inputs positive so the DEM clock stops (distortion). Then the DEM clock frequency doesn't matter as the DEM oscillator no longer works.
The DEM clock affects bit accuracy (you notice distortion when the DEM circuit doesn't work). So with a good DEM clock, this distortion can be minimized. When all is optimal one could achieve almost true 16 bit resolution (around 85 ... 95dB range). Most simple power amps can't cover this very large range, only a fraction of it. So the subtle differences remain inaudible.
Hi John, I see now. Many thanks for the explanation, it's nice to be able to understand what is happening.
I downloaded that test cd and have found that DEM is working, but only on 2.82MHz. I will test properly in the main system which is much higher resolution in a moment.
When I have some time off over christmas, I will have a proper read through that mega-thread of yours.
Thanks again, Lee.
I downloaded that test cd and have found that DEM is working, but only on 2.82MHz. I will test properly in the main system which is much higher resolution in a moment.
When I have some time off over christmas, I will have a proper read through that mega-thread of yours.
Thanks again, Lee.
Just reread your post Lucas, and noticed that you removed the TL431s...did you notice a difference in SQ?
Having one or two TDA1541As to hand, I bought some of Oliver's PCBs with a view to finishing off projects that had, well, taken a lengthy break . . .
I've just finished the first which uses a "crownless" 1541A chip and PSUs based on LM317s and 7805s in a Cascode arrangement (not the usual double regulator bodge) that I bought in kit form about ten years ago but never used. Input is via one of DD's USB to I2S modules and the I/V stage uses a couple of Sowter 8347 transformers that did duty for years in an upgraded CD player. Though it cost only buttons, it sounds rather good.
However, to get it to work I had to remove pull-up and pull-down resistors R28 and R31 in the BCK line. Until I did that, no sound at all.
Has anyone else tried DD's module in this application? If so, did you have the same trouble or have I done something wrong?
OTOH, if anyone does try the module and gets no sound, this might be worth a try.
Dave
I've just finished the first which uses a "crownless" 1541A chip and PSUs based on LM317s and 7805s in a Cascode arrangement (not the usual double regulator bodge) that I bought in kit form about ten years ago but never used. Input is via one of DD's USB to I2S modules and the I/V stage uses a couple of Sowter 8347 transformers that did duty for years in an upgraded CD player. Though it cost only buttons, it sounds rather good.
However, to get it to work I had to remove pull-up and pull-down resistors R28 and R31 in the BCK line. Until I did that, no sound at all.
Has anyone else tried DD's module in this application? If so, did you have the same trouble or have I done something wrong?
OTOH, if anyone does try the module and gets no sound, this might be worth a try.
Dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.