Thanks for your help Jan.
I'm using the digital input of the DCX2496 so I have to have the volume control between the DACs and the amps. Volume control on the digital side would cause lower dynamic range.
Most probably the best solution would be to use a AK4395 or AK4396 DAC with integrated digital volume control (doesn't reduce dynamic range). But that would mean to change the firmware of the DSP. Is anybody out there who can do this???
I'm using a motorized 6-gang pot from Alps. It's 15 Euro only (http://www.albs.de/). You also get the complete remote control from that source (60 Euro) but there might be cheaper solutions.
I optimized the pot with some external resistors, so that the gang errors are very low. The total resistance of the 6 different gangs still has an error of about +/-5% (7-8k).
So I would replace the 'positive' 10k resistor in your circuit with my 6 gang pot an the 'negative' 10k resistor with 6 trimmable resistors (about 6.5k fixed + 2 k trimmable). Is that OK?
I'm going to have a small 2U housing for the DCX2496 and 6 UcD 180/400 amps. I won't use the DCX housing any more including the front panel. So everything is in one box with very short cables.
Regards,
Frank
I'm using the digital input of the DCX2496 so I have to have the volume control between the DACs and the amps. Volume control on the digital side would cause lower dynamic range.
Most probably the best solution would be to use a AK4395 or AK4396 DAC with integrated digital volume control (doesn't reduce dynamic range). But that would mean to change the firmware of the DSP. Is anybody out there who can do this???
I'm using a motorized 6-gang pot from Alps. It's 15 Euro only (http://www.albs.de/). You also get the complete remote control from that source (60 Euro) but there might be cheaper solutions.
I optimized the pot with some external resistors, so that the gang errors are very low. The total resistance of the 6 different gangs still has an error of about +/-5% (7-8k).
So I would replace the 'positive' 10k resistor in your circuit with my 6 gang pot an the 'negative' 10k resistor with 6 trimmable resistors (about 6.5k fixed + 2 k trimmable). Is that OK?
I'm going to have a small 2U housing for the DCX2496 and 6 UcD 180/400 amps. I won't use the DCX housing any more including the front panel. So everything is in one box with very short cables.
Regards,
Frank
I believe the internal volume control inside those akm is also bit decimation. I am working with a coder on a volume control project for the dcx using wolfson wm8816 chips. They are about the closest thing to a ladder rack, but compact enough you can stick it inside the dcx chassis. It will have a remote display with a 20x2 lcd and rotary for interface. Hope to have it done mid-january. I would look for a better option than pot and the akm. Maybe the twisted pear audio boards would work for your application.
Hans
Hans
The clock backwards config. looks interesting,but as the DCX has a clock frekv. of 24,576Mhz and the CD has 16,9344 how to make it work?And as I have two more devices in the digital chain,a M-1000 Roland mixer and the DEQ2496.The M-1000 has both clock in and out,and the DEQ has clock input,that shouldn´t be a problem?Will they adapt to whatever the clock frekv.is?Would be fine if I could make this work..
It to said that S/PDIF can´t directly be made AES/EBU,but there are some chips that can do it,is this nessery?
Hi,
some of the bits in the data stream are deliberately different between the commercial AES/EBU and the user SP/DIF.
Fortunately some software ignores these particular bits and allow deciphering of the data. Just depends on what the manufacturer has opted to do.
The voltage levels are also different and easily overcome to make the two standards compatible.
some of the bits in the data stream are deliberately different between the commercial AES/EBU and the user SP/DIF.
Fortunately some software ignores these particular bits and allow deciphering of the data. Just depends on what the manufacturer has opted to do.
The voltage levels are also different and easily overcome to make the two standards compatible.
Hi Jan.
it's me again. I drew two circuits. The first shows DCX and UcD without pot and the second how I would integrate the pot.
I thought AC coupling caps are best after the pot because of the high input resistance of the instrumental amp of the UcD?
The Ucd amps have an on/off control input (http://www.diyaudio.com/forums/attachment.php?s=&postid=441385&stamp=1090523307) which is connected to the Hypex supply which has a power on/off circuit with power on delay. Might be this is already sufficient. Perhaps it is possible to connect the mute output of the DCX here too.
Do you have any complaints?
My crossover points are 127Hz and 1,35kHz with a 48dB slope. I've no possibility to calculate the filter. Is there a chance that you help me to optimize the filter?
Kind regards and best greetings from Bavaria,
Frank
it's me again. I drew two circuits. The first shows DCX and UcD without pot and the second how I would integrate the pot.
I thought AC coupling caps are best after the pot because of the high input resistance of the instrumental amp of the UcD?
The Ucd amps have an on/off control input (http://www.diyaudio.com/forums/attachment.php?s=&postid=441385&stamp=1090523307) which is connected to the Hypex supply which has a power on/off circuit with power on delay. Might be this is already sufficient. Perhaps it is possible to connect the mute output of the DCX here too.
Do you have any complaints?
My crossover points are 127Hz and 1,35kHz with a 48dB slope. I've no possibility to calculate the filter. Is there a chance that you help me to optimize the filter?
Kind regards and best greetings from Bavaria,
Frank
Attachments
Hi,
did my first SPICE today. The following values seem to be not too bad: First order filter R500, 1500pF, second order filter R250, 1000pF and AC coupling caps 2.2uF. See attached file (change appendix to .asc and use http://www.linear.com/designtools/softwareRegistration.jsp).
Has anybody the circuit of the analog and AES/EBU input buffers of the DCX2496? I'd like to get rid of the whole I/O buffer PCB.
Thanks,
Frank
did my first SPICE today. The following values seem to be not too bad: First order filter R500, 1500pF, second order filter R250, 1000pF and AC coupling caps 2.2uF. See attached file (change appendix to .asc and use http://www.linear.com/designtools/softwareRegistration.jsp).
Has anybody the circuit of the analog and AES/EBU input buffers of the DCX2496? I'd like to get rid of the whole I/O buffer PCB.
Thanks,
Frank
Just discovered that you the DEQ has the option to set the In/Out to S/PDIF OR AES/EBU,but the DCX only AES/EBU
Anyway,how to get 16,9344 from a 24,576Mhz clock,I now sommeone knows..😉
Thanks Ryssen for the analog IO section of the DCX2496.
Made some mod to the analog inputs too. See attached file (change appendix to .asc and use http://www.linear.com/designtools/softwareRegistration.jsp)
The LT1001 is caused by the LT library only. Use AD8620 or something of your taste.
C1 and C2 should be low dissipation caps (polyprop). With higher values of R1 and R2 (eg. 100k) values of C1 and C2 can be reduced (1 to 2.2 uF). R7 can be reduced for lower voltage inputs (higher gain). The three R332 and 1500pF parts are part of the DSP PCB!
Since this is my second analog circuit at all comments are welcome!
Regards,
Frank
Made some mod to the analog inputs too. See attached file (change appendix to .asc and use http://www.linear.com/designtools/softwareRegistration.jsp)
The LT1001 is caused by the LT library only. Use AD8620 or something of your taste.
C1 and C2 should be low dissipation caps (polyprop). With higher values of R1 and R2 (eg. 100k) values of C1 and C2 can be reduced (1 to 2.2 uF). R7 can be reduced for lower voltage inputs (higher gain). The three R332 and 1500pF parts are part of the DSP PCB!
Since this is my second analog circuit at all comments are welcome!
Regards,
Frank
oettle said:
Hi Frank,
Apologies for the late reply, missed your post!
This will certainly work, and the choice of the pot value seems sensible at first glance. The only remark I would have is that the pot setting influences the filter after the DAC; the lower the volume, the higher the filter cutoff frequency which would in theory give you more hf noise with lower volume. Then again, since this hf noise is high in frequency you probably don't hear it. I don't know how resistant the uCD is to hf noise?
This is probably best checked by just trying it out.
Jan Didden
Hi Jan,
I don't understand why you think the pot setting has influence on the filter.
Since the input resistance of the UcD is rather high (200k) the total resitance (10k) of the pot (which is the load of the filter) stays rather unchanched (about 9.5k at full volume).
So the pot setting has an influence on the filter since the total resistance of the pot varies between 10k and 9.5k but this minor change has a negligible influence on the filter (about 3% at 20kHz).
But perhaps I've overlooked something?
Regards,
Frank
I don't understand why you think the pot setting has influence on the filter.
Since the input resistance of the UcD is rather high (200k) the total resitance (10k) of the pot (which is the load of the filter) stays rather unchanched (about 9.5k at full volume).
So the pot setting has an influence on the filter since the total resistance of the pot varies between 10k and 9.5k but this minor change has a negligible influence on the filter (about 3% at 20kHz).
But perhaps I've overlooked something?
Regards,
Frank
Frank, I was replying to your circuit in post 826. The pot resistance changes the source resistance seen by the last filter capacitor (the one after the two 750 ohms resistors), so in theory the cap value should be recalculated (increased).
I now see that I misread the circuit also: I initially thought you used the pot as a variable resistance and that would have a large effect on the cap (and also vary the required cap value with pot setting). I now see that this is not the case, you use the pot as a real attenuator, so the effect is much less and also constant (provided the ucd has high Zin as you stated).
You are right, no changes are required.
Jan Didden
I now see that I misread the circuit also: I initially thought you used the pot as a variable resistance and that would have a large effect on the cap (and also vary the required cap value with pot setting). I now see that this is not the case, you use the pot as a real attenuator, so the effect is much less and also constant (provided the ucd has high Zin as you stated).
You are right, no changes are required.
Jan Didden
Hi Janneman,
are you sure you mis-read the schematic?
the RF filter Rs+100r & 100p are tied into the pot setting.
The pot setting effectively becomes Rs whenever the pot is NOT at maximum setting.
are you sure you mis-read the schematic?
the RF filter Rs+100r & 100p are tied into the pot setting.
The pot setting effectively becomes Rs whenever the pot is NOT at maximum setting.
Yes but the 100 ohms is effectively open ended with the very high ucd input Z, and the 100 ohms/ 100pF has a rather high cutoff point. There is influence but not too much. The pot itself increases the -3dB point of the filter and the 100 ohms/ 100 pF pulls it down a bit again. It could be tweaked but I think it's OK the way it is.
Jan Didden
Jan Didden
AndrewT said:
Andrew I missed your point, you are right. The pot gets in series with the 100 ohms and then the 100pF becomes significant! Definitely needs to be adapted.
Jan Didden
Hi,
tried hard to simulate what you are talking about. I can't see that the pot setting has any not negligable influence on the filter or attenuation.
The instrumental amplifier on the UcD PCB adds additional attenuation and phase shift at the high end. So the question is what is the optimal filter setting (attenuation at 20kHz and 40kHz).
Regards,
Frank
tried hard to simulate what you are talking about. I can't see that the pot setting has any not negligable influence on the filter or attenuation.
The instrumental amplifier on the UcD PCB adds additional attenuation and phase shift at the high end. So the question is what is the optimal filter setting (attenuation at 20kHz and 40kHz).
Regards,
Frank
Each part of the pot appears as the source Z for the 100 ohms and 100pF on the ucd input side. That means you have a lpf of (very roughly, there's more issues) of (part of the pot + 100 ohms) and 100pF. That is not inconsequential.
BTW at mid point of the pot your volume is zero, you know that of course?
Jan Didden
BTW at mid point of the pot your volume is zero, you know that of course?
Jan Didden
Hi,
if I'm right, the R100, 100 pF filter on the Ucd is a HF filter to attenuate noise caused by the wiring. The filter and the pot setting is negligible for frequencies below 20kHz.
The far more important question is what is the optimal (second order) filter setting for the tweeter channel (attenuation at 20 kHz, 40 kHz, 80 kHz). Might depend on whether you have CD or SACD data. Does anybody know about some theory?
Regards,
Frank
if I'm right, the R100, 100 pF filter on the Ucd is a HF filter to attenuate noise caused by the wiring. The filter and the pot setting is negligible for frequencies below 20kHz.
The far more important question is what is the optimal (second order) filter setting for the tweeter channel (attenuation at 20 kHz, 40 kHz, 80 kHz). Might depend on whether you have CD or SACD data. Does anybody know about some theory?
Regards,
Frank