I'm gonna buy me a LOT of de-solder wick 🙂
Never touching that solder-sucker again.
Used a lot och flux on the wick and ended up using a rather hot iron, 350C.
I guess I won't know the state of the IC's until I plug them in?
'
Never touching that solder-sucker again.
Used a lot och flux on the wick and ended up using a rather hot iron, 350C.
I guess I won't know the state of the IC's until I plug them in?
'
Last edited:
This came today, no pics of the content of the largest package though...
Mcap 8,2uF
Cross cap 120nF
ELNA silmic ROS 100uF/25V, 100nF WIMA MKP, 0,01uF WIMA FKP.
BYV27-200
Mcap 8,2uF
Cross cap 120nF
ELNA silmic ROS 100uF/25V, 100nF WIMA MKP, 0,01uF WIMA FKP.
BYV27-200
Made a few errors placing the decoupling caps yesterday, fixed these and soldered the four 8,2nF FKP's.
Cleaned the camera lens a bit, but I have still to learn how to use my fiances kamera. As you can see there's some flux that needs to be removed.
DAC progress looks good
Hi Mayday, Looks like your making good progress. I put the finishing touches on my 1st build today. I swapped out the caps on the MSB (most significant bit) pins 13, 18. I fit 0.75 ufd polypropylene there. I didn't have enough room to fit 1 ufd caps. What a difference with that and the DEM capacitor change. Doesn't sound like a TDA1541A anymore or any other DAC. 😀 Sounds more like a good vinyl rig. I ordered transformers for my second build "experimental unit". That will be very interesting. I'm doing something very different on that one.
Keep up the good work. Sorry I'm not able to post pics yet. Dave
Hi Mayday, Looks like your making good progress. I put the finishing touches on my 1st build today. I swapped out the caps on the MSB (most significant bit) pins 13, 18. I fit 0.75 ufd polypropylene there. I didn't have enough room to fit 1 ufd caps. What a difference with that and the DEM capacitor change. Doesn't sound like a TDA1541A anymore or any other DAC. 😀 Sounds more like a good vinyl rig. I ordered transformers for my second build "experimental unit". That will be very interesting. I'm doing something very different on that one.
Keep up the good work. Sorry I'm not able to post pics yet. Dave
Thanks Dave 🙂
I'm glad your happy with the sound!
I used polypropylene(MKP, and a few FKP) caps for all decoupling caps, that's why I had to put a few of them under the pcb. It was hard to find all values according to UV's PDF(1uF, 220nF, 120nF, 33nF, 33nF, 8,2nF, 8,2nF all x2) from one maker, so while all are polypropylene brands are WIMA, RIFA and Jantzen.
All filmcaps on the PCB is polypropylene actually, I don't know it that will effect the final result but I hope so.
DEM cap is 122pF <%1 Silver Mica. Used an identical cap in the CD-60 w great results.
Btw, if you're using AD797, and haven't put a 100R resistor between -IN and OUT(have to cut trace between the pins too) the AD797 might oscillate. This is only for the second pair, the buffer. COULD this be the cause for the clicking sound some report?
I'm glad your happy with the sound!
I used polypropylene(MKP, and a few FKP) caps for all decoupling caps, that's why I had to put a few of them under the pcb. It was hard to find all values according to UV's PDF(1uF, 220nF, 120nF, 33nF, 33nF, 8,2nF, 8,2nF all x2) from one maker, so while all are polypropylene brands are WIMA, RIFA and Jantzen.
All filmcaps on the PCB is polypropylene actually, I don't know it that will effect the final result but I hope so.
DEM cap is 122pF <%1 Silver Mica. Used an identical cap in the CD-60 w great results.
Btw, if you're using AD797, and haven't put a 100R resistor between -IN and OUT(have to cut trace between the pins too) the AD797 might oscillate. This is only for the second pair, the buffer. COULD this be the cause for the clicking sound some report?
the clicking is due to clocking issues, nothing to do with the IV stage/buffer analogue domain
Last edited:
What is the best option? Slaving or mastering CS8412?
Will use 0,3ppm TCXO, will this help the issue?
IF the clicking appears in my dac when done, would a DIR9001 on a CS8412 adapter solve the problem?
Clicking sound
Hi Mayday, No, I am not using the analog stage at all on the Analogmetric DAC. I take the output to an OPA603 as I/V (I/U) , a 7th order GIC Bessel filter and an OPA627 output buffer. Yes, the AD797 can oscillate. I know you plan to use a Tube stage and I recommend that. I agree with qusp, the clicking is related to overrun/ underrun of the incoming SPDIF. Might be very hard to fix that. One thought is to build a large FIFO (first in first out) and output clock the data. On my build I was very lucky and I don't get clicks unless I'm inputting 48 KHZ data and then it's not to often. My second build did have the problem. So my plan is to freshen up the old Magnavox CDB650 and add a clock input. I will run the CDB650 as a slave to the DAC. Since I use a Hagclock I also have the option of adding a small variable cap in the tank circuit to adjust the frequency of the clock. Now I suppose I could try something fancy like recovering the clock from the input receiver and using that to drive the Hagclock up and down as needed. The question then is why use the masterclock at all. I think the "slave transport" is probably the easy solution. Otherwise, if the problem is "a problem" on your DAC you could simply use the CS8412 as master and use the recovered clock to drive the DAC. I guess I'd wait and see if it's a problem on your DAC. I do have another idea however it wouldn't be that easy to do using a DDS oscillator. Dave
Hi Mayday, No, I am not using the analog stage at all on the Analogmetric DAC. I take the output to an OPA603 as I/V (I/U) , a 7th order GIC Bessel filter and an OPA627 output buffer. Yes, the AD797 can oscillate. I know you plan to use a Tube stage and I recommend that. I agree with qusp, the clicking is related to overrun/ underrun of the incoming SPDIF. Might be very hard to fix that. One thought is to build a large FIFO (first in first out) and output clock the data. On my build I was very lucky and I don't get clicks unless I'm inputting 48 KHZ data and then it's not to often. My second build did have the problem. So my plan is to freshen up the old Magnavox CDB650 and add a clock input. I will run the CDB650 as a slave to the DAC. Since I use a Hagclock I also have the option of adding a small variable cap in the tank circuit to adjust the frequency of the clock. Now I suppose I could try something fancy like recovering the clock from the input receiver and using that to drive the Hagclock up and down as needed. The question then is why use the masterclock at all. I think the "slave transport" is probably the easy solution. Otherwise, if the problem is "a problem" on your DAC you could simply use the CS8412 as master and use the recovered clock to drive the DAC. I guess I'd wait and see if it's a problem on your DAC. I do have another idea however it wouldn't be that easy to do using a DDS oscillator. Dave
I don't think the DIR9001 can solve this problem. It might have better jitter performance, that's the only benefit that I can see over the CS8412.
Ok, I'll cross that bridge if/when I get there then 🙂
I've got all I need for tube output except 50R I/V resistors, 56r 50W resistor(bring down 24Vdc to 6Vdc)and the 24V/1A supply(cheap ebay wallwart). A friend gave me the matched 2SK170(5,4mA Idss) and I've traded stuff I didn't need for the rest.
I've got all I need for tube output except 50R I/V resistors, 56r 50W resistor(bring down 24Vdc to 6Vdc)and the 24V/1A supply(cheap ebay wallwart). A friend gave me the matched 2SK170(5,4mA Idss) and I've traded stuff I didn't need for the rest.
Don't know if it has any relevance at all, but source will be a SB Classic.
Clock on DAC will be Vanguard 0,3ppm 11.2896Mhz TCXO.
Clock on DAC will be Vanguard 0,3ppm 11.2896Mhz TCXO.
Has EVERYTHING arrived in the post yet ?
Just wondering...
Just catching up with the posts - been away
DAC looks fantastic btw - not far to go by the look of it.
You've been very busy and now.... a convert to solder wick - much easier isn't it ?😉
Just wondering...
Just catching up with the posts - been away
DAC looks fantastic btw - not far to go by the look of it.
You've been very busy and now.... a convert to solder wick - much easier isn't it ?😉
Thanks, I hope it sounds fantastic too 🙂
No, the can you sent is still to arrive.
So is DV709, Rubycon ZA, CS8412, last of the regs, heatsinks and isolationpads and quite a lot of solder wick 😉
It is as you said, removes enough solder for the IC's to drop of the pcb. Easier, faster and saves the PCB.
Last edited:
Hi Mayday, Just so you know the input for the SPDIF is not quite right on the Analogmetric DAC. The best fix I could come up with is to move the 75 Ohm teminating resistor (meant to terminate the coax) to across the green screw terminals. I soldered mine under the board where the SPDIF goes in. Don't just add another 75 Ohm resistor. I see you plan to use the DV709. This will work just fine. Take a look at the datasheet for the CS8412. Analogmetric uses the right parts just in an odd way. 🙂 Dave It's important to properly terminate the coaxial input for low reflection and jitter performance
The 75R has been changed in the BOM I got to 76,8R.
So, it should be move to across pins for SPDIF connector even with the changed value?
Didn't even realize the DV709 was optional lol!
So, it should be move to across pins for SPDIF connector even with the changed value?
Didn't even realize the DV709 was optional lol!
Input receiver
Hi Mayday, The value you mentioned is pretty close and should work OK. The point of the resistor is to terminate the coaxial cable. That way the SPDIF signal won't be reflected back to the source. In an ideal world I would use whatever value matches your intended cable. I use a silver Teflon cable with 75 Ohm impedance. Matching it reduces jitter. Yes, the transformer could be jumpered. I ran mine that way for a while. I finally relented for galvanic isolation reasons to run it. I've heard no difference. I sourced my transformer from Digikey so I run a different equivalent part. Hope that helps...😀 Dave
Hi Mayday, The value you mentioned is pretty close and should work OK. The point of the resistor is to terminate the coaxial cable. That way the SPDIF signal won't be reflected back to the source. In an ideal world I would use whatever value matches your intended cable. I use a silver Teflon cable with 75 Ohm impedance. Matching it reduces jitter. Yes, the transformer could be jumpered. I ran mine that way for a while. I finally relented for galvanic isolation reasons to run it. I've heard no difference. I sourced my transformer from Digikey so I run a different equivalent part. Hope that helps...😀 Dave