Tonewise upgrading the SY77 to SY99 specs

[knolan]

Thanks!

Absolutely no hurry at all - even in a months will be fine (I'm slow at moving on these things, though always get there ).

[Mooger5]

SY77 with NE5532. The difference to the 4558 is minimal. The 5532 sounds cleaner and with the 4558 I think I can hear some distortion, but that could be because I read datasheets and related literature.

I took some pictures during the swapping which I´ll post tomorrow with a few comments.
This reminds that the present topic is nothing special, no major feat; just opamp swapping and a lot of talk

[Mooger5]

Happy to make a donation via paypal by the way.

I´ve always refused money offers in exchange for help, but you don´t seem like a starving musician or a jobless student, so just do what you think is best, man...

pics coming tomorrow

[Mooger5]

morrow

[knolan]

Hi Mooger5 -

If you can bring these tests to a firm conclusion and can confidently say that a particular modification will affect an improvement - verified by tests - and can then provide a detailed description of the procedure, parts list, etc - then I would be in a position consider implementing it.

If you feel you can't bring that clarity to it - absolutely no issue at all - i accept you're doing this for your motivation and your post here are intriguing, and appreciated. I'm just emphasising that for me to affect one of my instruments with any such improvement, I'd need to be able to present to the engineer a detailed parts list, description and so on (and if you can produce that; I'd be willing to make a contribution via paypal).

But as said - no issue at all if that's not your thing - as said - you're work and posts as is are quite fascinating.

Best regards,
Kevin.

[Mooger5]

Hi Kevin, thanks for the heads-up.
The M5238s are due to arrive on September the 6th according to the tracker. Hopefully earlier but anyway we´ll have to wait a bit more. I also ordered PCM61 DAC chips, supposedly 18-bit drop-in replacements for the 16-bit PCM56. Should be interesting...
I´ll have to make all experiments on the TG77, much more practical.
Planning to add the second filter stage on a perfboard and leave it at that in case of a noticeable improvement, or otherwise remove it afterwards. Intriguing part, simulations suggest different component values so thy might have tweaked them to get rid of some ringing or whatever. Something definitely going on there or they wouldn´t go to the extent of redesigning/rebuilding an entire PCB just to fit in such a small part.

The photos are too dark. I´ll post one or two to illustrate some pratical advices and the rest I´ll have to grab from the internet.

Cheers.

[Mooger5]

Here´s a simple but revealing test: I bypassed the 2nd order Sallen-Key low pass filter in the TG77 on both channels of Output 1 and recorded one of the factory demos ("Power Play") just to see how it went. At first it sounds OK. Take special attention to the horn stabs and the guitar solo...



Now the same song with the filter ON:



Sounds alive in comparision, right? Now imagine what the steeper 4th order filter of the SY99 could do here

[Mooger5]

Just leaving this here. Bye.

[Mooger5]

C12 above should be 2nf instead of 100nf.

Here´s waiting for the M5238s to arrive:

[Mooger5]

Damn cheap components. R5 is reading 3.3k. And it´s because it IS a 3k3 resistor! It´s the first two orange stripes that look almost red. I think it´s possible to notice in the photo. Glad I detected it in time.

About the signal capacitors, you don´t see those values often and I couldn´t get them locally, so I used what I already had. The typical values that are closer to 6.2nf and 2nf are 5.6nf (the 6n8 PEs that I have here all read >7nf) and 2.2nf ...

[Mooger5]

Board and chips installed. Fingers crossed.

[Mooger5]

Something´s not right for the moment. Sound is clean but very faint. In Multi mode volume is normal for about five seconds and goes faint. Changing to another multi sound the volume gets normal again and then drops.
There are pops when powerinf on and off, so there is DC is the audio path. I did put the filter before the electrolytic caps, the non polarized ones added to the new board. Maybe some wires are misplaced. Gonna check everything now.

[Mooger5]

Think I found the culprits: as the stripboard tracks where the blocking caps go had to be cut, a couple tracks resulted in hairline solder bridges bypassing the capacitors. Poor eyesight just leads to this even after using the "official" Vero tool.
Here you can see one of the badly cut tracks, next to a good one for reference. Everything else checks fine, so I think this is it.

[Mooger5]

It works and it sounds better, I think



Let me know what you think|

[Mooger5]

Thank you for your kind comment on the Soundcloud page madtheory.
The m5238 chip is something else, right? And the aid of a steeper, ultrasonic LP filter seems counterintuitive but does improve resolution and HF detail. It's these enhanced highs that provide the clearer spatial clues IMO. This Aquavox preset has auto panning on the FM and the PCM elements.
I'm baffled by how much busier left/right they are now. The TL072 without the extra filter sounds almost in mono by comparision.