Re: Re: Re: Single Side PCB PDFs for VREF
Just thought of another thing as I am about to install Vref. I made up the board complete and will mount this on a pillar very close to C6, with short lead in and outs.
If I connect + and Gr in from VA and Gr to the Vref, should I then connect +out and Gr from Vref to C6 and R13 Gr?, or just connect Vref to C6? Presumably it is best to avoid any ground loops as this can cause as much sonic change as installing Vref by generating noise..
Any thought? How did you do it? Am I too cautious?
Fred
Russ White said:
Just thought of another thing as I am about to install Vref. I made up the board complete and will mount this on a pillar very close to C6, with short lead in and outs.
If I connect + and Gr in from VA and Gr to the Vref, should I then connect +out and Gr from Vref to C6 and R13 Gr?, or just connect Vref to C6? Presumably it is best to avoid any ground loops as this can cause as much sonic change as installing Vref by generating noise..
Any thought? How did you do it? Am I too cautious?
Fred
Re: Re: Re: Re: Single Side PCB PDFs for VREF
Hi Fred. It's a good question, the answer is you only want to run VREF to C6. The GND should be run only to VA GND.
So you will have a total of three wires out. One to VA, one to GND, and one to the VREF input of the AVCC supply (opamp).
Have fun!
Cheers!
Russ
fmak said:
Hi Fred. It's a good question, the answer is you only want to run VREF to C6. The GND should be run only to VA GND.
So you will have a total of three wires out. One to VA, one to GND, and one to the VREF input of the AVCC supply (opamp).
Have fun!
Cheers!
Russ
Re: Re: Re: Re: Re: Single Side PCB PDFs for VREF
I have installed Vref and it makes a big difference to 'musicality' which I lost somewhat when I started to stack the units
. Very nice. I also installed small ferrite rings to the Vref feed and the output wire.
Russ, I have found a small 22.1kHz modulation of all the supplies from the Buffalo/Ivy, with or without Vref. Any idea as to what this may be due to?
I shall try to establish the absolute level as I am currently amplifying the signal from the PSs.
Happy New Year
Fred
Russ White said:
I have installed Vref and it makes a big difference to 'musicality' which I lost somewhat when I started to stack the units
. Very nice. I also installed small ferrite rings to the Vref feed and the output wire.
Russ, I have found a small 22.1kHz modulation of all the supplies from the Buffalo/Ivy, with or without Vref. Any idea as to what this may be due to?
I shall try to establish the absolute level as I am currently amplifying the signal from the PSs.
Happy New Year
Fred
Fred.
Have you tried moving the DAC away from the IVY ?
See my post, http://www.diyaudio.com/forums/showthread.php?postid=1696262#post1696262
It made all the difference one could ask for.
Have you tried moving the DAC away from the IVY ?
See my post, http://www.diyaudio.com/forums/showthread.php?postid=1696262#post1696262
It made all the difference one could ask for.
avr300 said:
yes, I started out with a flat configuration. Now I am trying to save space.
What do you think of the active Vref? Wonder what if one restores the filter to the IC?
The Vref mod is a keeper, surely.
But the increased clearance is even more important.
What I was suggesting was simply to see, whether the 22k modulation decreases when the space between the two PCB's increases.
But the increased clearance is even more important.
What I was suggesting was simply to see, whether the 22k modulation decreases when the space between the two PCB's increases.
avr300 said:
Thanks for reminding me. Now that I am prompted, I don't think I was there when the modules were laid flat. Also, could be the Mux unit which wasn't there before.
Will investigate.
Russ any thoughts?
ps: I really like the Vreg
Hi Fred.
I don't have the modulation you seem to have. Perhaps a wiring or a stacking issue as has been suggested.
I am really not sure.
Glad its working though.
Cheers!
Russ
I don't have the modulation you seem to have. Perhaps a wiring or a stacking issue as has been suggested.
I am really not sure.
Glad its working though.
Cheers!
Russ
Russ White said:
Cracked it! It's a stacking issue even though I had a ground plane between the two boards. Could even be the ground plane!
Probing with a scope probe 2 cm below the Buffalo and above the plane, I found different freq pickups at different points. The area below the ESS chip is very energetic with strong clipped waves; the area under the PIC emits around 21k and other areas give out 38k or so!.
Only the 21k emission is embedded into power supply lines.
So more work, separate the modules again!!!
Russ and others, please try it and see what you get.
fmak said:
I don't actually stack the IVY with the buffalo anymore. Mostly because I am working on a couple new output stages. 🙂 They are mono- so they go either side of the DAC.
But when I do I generally put the Buffalo on top with 1" standoffs.
I am not picking up any EMI.
BTW the PIC is normally idle (powered down) except right at start up. So I would not expect it to be emitting anything.
Glad it seems to be better now for you.
Cheers!
Russ
Fred.
Do one more test. See if the noise correlates with the DAC being locked onto a digital stream.
I only had the noise problem when the DAC was locked. When unconnected, and NOT automuted there was no noise.
Do one more test. See if the noise correlates with the DAC being locked onto a digital stream.
I only had the noise problem when the DAC was locked. When unconnected, and NOT automuted there was no noise.
avr300 said:
It looks like the ground plane I put between the Buffalo and Ivy, Removed it but there is still about 30 uV of modulated noise on the supply lines. It's small but my alw regulators should not have this. will investigate.
Grrrrr...
I went and made some PCBs for the CCS - and all went swimmingly well, got a lovely panel of 4 boards:
Really happy with it.
Until I went to stuff the board. I've made a right balls of it somehow. It seems that what I have is the mirror of what I should have ie it doesn't match the silkscreen.
I used the toner method - I definitely printed the bottom_mirror file because I've gone back and checked that against Russ's PDF. What the hell have I done wrong?
Heres another pic of the half stuffed board. See the way the 3 resistors R1,4 and 5 should be towards the bottom left hand corner but they are in the left top corner.
So what I'm wondering is.... If I just flip the board over like this:
And solder the components to this side (ie the Wrong side) then wouldn't it be OK?
Help!!
I went and made some PCBs for the CCS - and all went swimmingly well, got a lovely panel of 4 boards:
An externally hosted image should be here but it was not working when we last tested it.
Really happy with it.
Until I went to stuff the board. I've made a right balls of it somehow. It seems that what I have is the mirror of what I should have ie it doesn't match the silkscreen.
I used the toner method - I definitely printed the bottom_mirror file because I've gone back and checked that against Russ's PDF. What the hell have I done wrong?
Heres another pic of the half stuffed board. See the way the 3 resistors R1,4 and 5 should be towards the bottom left hand corner but they are in the left top corner.
An externally hosted image should be here but it was not working when we last tested it.
So what I'm wondering is.... If I just flip the board over like this:
An externally hosted image should be here but it was not working when we last tested it.
And solder the components to this side (ie the Wrong side) then wouldn't it be OK?
Help!!
woodturner-fran said:
I guess you missed this post before you made your boards. 🙂
http://www.diyaudio.com/forums/showthread.php?postid=1695593#post1695593
What you have should be ok used as a top side as long as you can make good solder joints. 🙂
Jut s test separately for the correct reference voltage prior to using for the buffalo.
Cheers!
Russ
Cheers!
You know I actually read that post, but somehow my brain took it up to use the mirrored side for toner method!
Anyway, all's not lost and I'll remount the stuff on the other side tomorrow and test it with a bench supply first before installing it. It should all go together ok I think. A good thing is that the toner method worked out really well for this which opens up the door for small PCBs in the future. Getting to grips with the likes of eagle isn't easy though I think. I had a go at it in the past but couldn't make head nor tail of it.
Thanks Russ for the help and suggestions. You were of course correct in your post linked above - it does all become clearer when you do it!
Fran
Anyway, all's not lost and I'll remount the stuff on the other side tomorrow and test it with a bench supply first before installing it. It should all go together ok I think. A good thing is that the toner method worked out really well for this which opens up the door for small PCBs in the future. Getting to grips with the likes of eagle isn't easy though I think. I had a go at it in the past but couldn't make head nor tail of it.
Thanks Russ for the help and suggestions. You were of course correct in your post linked above - it does all become clearer when you do it!
Fran
Nope, we ain't working.
R1 is 4.8k
R2 is 1K
R3 is variable - tried from 6.2k to 47k
R4 and R5 is 221R
C1 and C2 are 100uF
Q1 is 2N5087 inserted opposite way to silkscreen
D1-3 are all HLMP6000 red LED with Vf of about 1.6-1.7V. all light up and glow with what I'd call normal brightness.
So what I'm seeing is very small adjustment range of Vref. I adjust R3 from 6.2 to 47K and all I get is from 2.6 to 3.08V and going from R3 10K to 47K only changes it from 2.85 to 3.08V.
Input voltage is 7.5V from a bench supply.
Any suggestions?
Fran
R1 is 4.8k
R2 is 1K
R3 is variable - tried from 6.2k to 47k
R4 and R5 is 221R
C1 and C2 are 100uF
Q1 is 2N5087 inserted opposite way to silkscreen
D1-3 are all HLMP6000 red LED with Vf of about 1.6-1.7V. all light up and glow with what I'd call normal brightness.
So what I'm seeing is very small adjustment range of Vref. I adjust R3 from 6.2 to 47K and all I get is from 2.6 to 3.08V and going from R3 10K to 47K only changes it from 2.85 to 3.08V.
Input voltage is 7.5V from a bench supply.
Any suggestions?
Fran
OK,
I think I have it fixed. I went back and made the "old" circuit with just the 2 1K resistors as shown on pg 42 and I got the same result. So I came to the conclusion that it was something to do with the HLMP6000 LEDs. Too low a Vf? Anyway, I swapped in some no-name el cheapo LEDs that I bought from ebay and right away I get 3.304V. So its looking good. That was with R3=10K. So I think I will try it out in the buffalo.
Should I expect some sag in that supply and if so what can I get away with?
Fran
I think I have it fixed. I went back and made the "old" circuit with just the 2 1K resistors as shown on pg 42 and I got the same result. So I came to the conclusion that it was something to do with the HLMP6000 LEDs. Too low a Vf? Anyway, I swapped in some no-name el cheapo LEDs that I bought from ebay and right away I get 3.304V. So its looking good. That was with R3=10K. So I think I will try it out in the buffalo.
Should I expect some sag in that supply and if so what can I get away with?
Fran
If you have 3.3V with 10K you can try just increasing or omitting R3. 🙂
If you need to you can also reduce R4 (I would not go lower than 100R or so) which will send more current through the LEDs which will raise the Vf a bit. 🙂
You can tune the output voltage either or both ways.
Red LEDs tend to have lower Vf than Green. I used Green.
Cheers!
Russ
If you need to you can also reduce R4 (I would not go lower than 100R or so) which will send more current through the LEDs which will raise the Vf a bit. 🙂
You can tune the output voltage either or both ways.
Red LEDs tend to have lower Vf than Green. I used Green.
Cheers!
Russ
Vref Again
Russ
Why have you left out the lf filter before the LM4562 on the board in feeding Vref? Is there a drive issue?
Seems it may be useful to reinstate lf filtering to a low freq.
Gone back to my old setup with ESS well away from PSs. Bass is a little 'plumb' .
Russ
Why have you left out the lf filter before the LM4562 on the board in feeding Vref? Is there a drive issue?
Seems it may be useful to reinstate lf filtering to a low freq.
Gone back to my old setup with ESS well away from PSs. Bass is a little 'plumb' .
- Status
- Not open for further replies.