[b0bb]

Everything in life is a compromise b0bb I'm not a genius of electronics like you.

Underhanded compliments like this is not a good thing if you are seeking help

 

[Xoverman]

For thos who are interested

 

[piaseczek]

For thos who are interested

What new have you discovered?

 

[Xoverman]

What new have you discovered?

I thought it might be a good idea to give analog/digital core power supply a more direct path, so that noise doesn't travel so much through the shared ground plain. And it turned out to be a good idea. It also gave me the opportunity to use a physically bigger inductor.

 

[Monolithic]

Yes, that is one of the main change LKS made when they rolled out the 3.2k bugfix version of the 004.
I would get rid of the LP38798, I am using the Bellesion SPLV78 outputting 1.3V

@b0bb - What series FP cap did you use for the 47uf? RNS or RNU? Is the 68uf RR7 too much output capacitance for the Belleson regulator?

Thanks.

 

[b0bb]

@b0bb - What series FP cap did you use for the 47uf? RNS or RNU? Is the 68uf RR7 too much output capacitance for the Belleson regulator?

Thanks.

These are RNU which was what I had on-hand. 68uF RR7 looks reasonable, as long as it fits in the space.

 

[piaseczek]

Those are the 1.1A @ 120mOhms that I originally listed.

They are 60% the height of the original inductors (0.5mm vs 0.9mm). Also take note that the datasheet states in Section 11.4:
"* Avoid a direct contact of the tip of the soldering iron with the product. Such a direct contact may cause cracks in the ceramic body due to thermal shock."

They also recommend pre-heating @ 150 deg C for one minute before soldering and less than 3 second soldering time.

Without hot air, your best bet would be to clean both pads, add solder to one pad, using a small tip heat the pad with solder at the edge of the pad and bring the inductor into the solder with tweezers, then remove the soldering tip and heat the other pad at the edge as you bring in solder to melt on the second pad, avoiding contact with the inductor. Solder should not rise above the top edge of the inductor.

I used a pre-heater for the board and hot air. I did not change the 5th inductor at the OCXO because I use a Pulsar clock that operates within the 200ma current limit. In the future, I may change the secondary regulator out altogether to increase the current available to the clock and would make any auxiliary component changes at that time.

If the low smd chip height is a concern, an alternative might be the LQM21PHR47NG0D which is 0.9mm height @ 1.3A @ 94mOhms.

Monolithic,
Have you replaced the inductors in your DAC? Has it improved the sound or is it just a minor upgrade?

 

[XVampireX]

I'll be selling my MH-DA004 soon enough, if anyone is interested

 

[Monolithic]

Monolithic,
Have you replaced the inductors in your DAC? Has it improved the sound or is it just a minor upgrade?

I replaced the inductors but didn't do an A/B comparison prior to upgrading the power supply section.

 

[piaseczek]

B0bb,
I have a problem with the screw on the right side of a photo, do you know if it could be disconnected?

 

[piaseczek]

I replaced the inductors but didn't do an A/B comparison prior to upgrading the power supply section.

The inductors are in my opinion the most difficult part to replace with the soldering iron. I will leave it as it is and only replace LM317s to LDOs + new FPCAPS. I hope it will be a good compromise.

 

[b0bb]

B0bb,
I have a problem with the screw on the right side of a photo, do you know if it could be disconnected?

Tha screw ties your DAC's groundplane to the chassis, replace the screw.
Not sure what type of thread LKS used, bring it to a hardware store to get the right replacement.

 

[piaseczek]

B0bb,
Will there be any problem if I replace all four nichicon KGs in the digital PSUs to the same value 22k uF 25V CDE 380LXs?
https://pl.mouser.com/ProductDetail...80LX223M025J052?qs=Yar5Zvc6MhVVTvnR4qZ%2BCw==

I'm guessing that clock will be happy with higher capacitance in the PSU (and higher PSRR).

 

[b0bb]

B0bb,
Will there be any problem if I replace all four nichicon KGs in the digital PSUs to the same value 22k uF 25V CDE 380LXs?
https://pl.mouser.com/ProductDetail/Cornell-Dubilier-CDE/380LX223M025J052?qs=Yar5Zvc6MhVVTvnR4qZ%2BCw==

I'm guessing that clock will be happy with higher capacitance in the PSU (and higher PSRR).

You will have to check if the caps can fit in the space

 

[piaseczek]

You will have to check if the caps can fit in the space

The caps will fit but is such o high capacitance (and 6.1A) recommended for the low power clock?