My_Ref Fremen Edition - Beta build/Fine tuning

[ClaveFremen]

Originally Posted by linuxguru

That's a typical symptom of either a dry (non-ohmic) joint on one of the resistor leads, or a flaky resistor with a cracked/loose/non-ohmic end-contact. Reversing it probably fixed the dry joint or made the end-contact sit snugly, thus solving the problem

Hi Siva,

all through hole signal resistors are socketed (see attached magnification) so the difference is not soldering related.

I've reversed it several times with always the same results.

Takman REX, in my experience, are directional.

The Same for PRPs.

Originally Posted by linuxguru

It could also account for some of the hum/crackling artefacts reported - in my experience, the MyRef is sensitive to dry joints at several location

This is a possible hint, maybe the problem is socket related...

[soongsc]

Originally Posted by ClaveFremen

Hi Siva,

all through hole signal resistors are socketed (see attached magnification) so the difference is not soldering related.

I've reversed it several times with always the same results.

Takman REX, in my experience, are directional.

The Same for PRPs.



This is a possible hint, maybe the problem is socket related...

Does it make sense to put two in opposite directions to make the result non-directional?

[ClaveFremen]

Originally Posted by soongsc

Does it make sense to put two in opposite directions to make the result non-directional?

Maybe...

I think that I've finally found where is the problem... LM317's power dissipation.

When idle temperature on LM337 is 44 °C, on LM317 57°C (23°C room temperature).

I've replaced both LM317, the new ones heatsinked, problem seems gone.

What puzzles me is that when on breadboard this problem did not appear.

Just for the sake of precision I've measured also both unregulated rails (+33,6Vdc and -34Vdc with 2x24Vac transformers) so there is not a voltage difference...

So for all beta builders: buy TO-220 thin heatsinks for LM317s

This Mouser code should be suitable: 532-576802B00

[squalor]

What puzzles me is that when on breadboard this problem did not appear

what changed ?

How many 68R and 270R should one buy to get a good match ? ... or can I pay PartsConnexion to match for me ?
Interesting heat sink on your 3886.
Are the Mundorf smoothing caps bigger (taller) than Panasonic TS-HA 50V at C101.201 ? Your M-Lytic are 10000uf 63V MLGO right. I think the chassis I am working on is too small.

[ClaveFremen]

Originally Posted by squalor

what changed ?

Maybe the fact that on breadboard there were more air around them... I don't know.

Originally Posted by squalor

How many 68R and 270R should one buy to get a good match ? ... or can I pay PartsConnexion to match for me ?
Interesting heat sink on your 3886.

I don't think matching is necessary with 1% resistors.

The 'heatsinks' are server heatsinks for harddisks... temporary, obviously...

Originally Posted by squalor

Are the Mundorf smoothing caps bigger (taller) than Panasonic TS-HA 50V at C101.201 ? Your M-Lytic are 10000uf 63V MLGO right. I think the chassis I am working on is too small.

They're both 50mm tall.

[soongsc]

Originally Posted by ClaveFremen

Maybe...

I think that I've finally found where is the problem... LM317's power dissipation.

When idle temperature on LM337 is 44 °C, on LM317 57°C (23°C room temperature).

I've replaced both LM317, the new ones heatsinked, problem seems gone.

What puzzles me is that when on breadboard this problem did not appear.

Just for the sake of precision I've measured also both unregulated rails (+33,6Vdc and -34Vdc with 2x24Vac transformers) so there is not a voltage difference...

So for all beta builders: buy TO-220 thin heatsinks for LM317s

This Mouser code should be suitable: 532-576802B00

How would you describe the sonic difference compared against your original breadboard mods? Is there anything about the sound that specifically catches your attention?

[ClaveFremen]

Originally Posted by soongsc

Is there anything about the sound that specifically catches your attention?

It's like a sort of ringing/distortion that occasionally comes out on level variations.

I think the problem is not PCB related but to the fact that the PS components, probably, don't like the resistor swapping and get damaged.

Just to be sure that it's not related to the amp section I've pulled out the entire PS and mounted the stock My_Ref PS, it sounds perfect.

After that I've mounted again the shunt transistors and resistors, it seems to sound perfect (but better ) too.

I've run out of LM317/LM337, in the next days I'll remount also them.

Now that I have all components I've mounted also the voltage limiter, it seems it has no adverse effects.

[ClaveFremen]

Today, I've mounted 14V BOM's zeners and so far it continues to sound good.

Just a note: while 1N53xx sounds fuller they, maybe have a touch of hardness but as hours are passing it seems to lower.

According to tracking packages left Italy this friday so I think that between the end of the next week and the beginning of the next should arrive at beta testers' home.

[superfigone]

You can solder the pins on the PCB to avoid slaughtering of the card

[ClaveFremen]

Originally Posted by superfigone

You can solder the pins on the PCB to avoid slaughtering of the card

If you're referring to sockets for easy swapping of components I do already use them