Personally ,i don't think you'll hear a huge difference....if you do it's more likely to be the fresh solder on those particular joints after the replacement.
Pity you can't improve on the op-amps.
That's where you'll hear a huge difference.

Also look at supply rail noise/regulation/tracking and symmetry. :eek:

Hi Chris,

You should'nt worry too much about the non-polarized electrolytics in your signal path, as long as they're decent quality. Stuff like SSL, Calrec, and so on are all made this way.

One thing you maybe could consider, is changing the C101/102 10u/16V electrolytics at the inputs of your tape returns to something like 1u or 2u2/50V polyesters. This appears to be the most critical point in the circuit - Both SSL and Calrec uses polyesters here. As they're loaded pretty weakly - with 22 KOhms each - this should give you plenty low-end response (Low corner ~3Hz @ 2u2, 6Hz @ 1uF - for sure better than what your taperecorder will reproduce)

Maybe also C117/118 in the insert returns, if you plan to use these as tape returns..

BTW: what board is this? I like the simple EQ design! And what are the IC's designated U101 and U105, they're unreadable at the scanned schematic..?

Jakob E.

I agree. Far too much importance is placed on caps in the signal path IMHO.

With the exception of replacing old electrolytics which may have gone bad over time I don't think you will hear any difference.

But still if people think its worth it then fine it certainly won't hurt (apart from your pocket!)

Originally posted by Jakob Erland:

Hi Chris,
One thing you maybe could consider, is changing the C101/102 10u/16V electrolytics at the inputs of your tape returns to something like 1u or 2u2/50V polyesters. This appears to be the most critical point in the circuit - Both SSL and Calrec uses polyesters here. As they're loaded pretty weakly - with 22 KOhms each - this should give you plenty low-end response (Low corner ~3Hz @ 2u2, 6Hz @ 1uF - for sure better than what your taperecorder will reproduce)

Hi,

Thanks for all the input, I think I will replace those caps that you mentioned with Polyesters.

U101 is a NJM2041SD
U105 is a RMNZ8178

whatever they are. I might like to improve U101 with something better, it's just a PITA to desolder. I was thinking about a more complete opamp upgrade but thought that it was probably overdoing it on this desk.

Also i've been looking around for a while for dual 100k rev log pots to no avail. Today I found a solution for those in Australia - Dick Smith has 100k dual log pots that can be pulled apart nicely for AU$1.15 each - a bargain!

Cheers

Chris

Originally posted by Jakob Erland:

BTW: what board is this? I like the simple EQ design! And what are the IC's designated U101 and U105, they're unreadable at the scanned schematic..?

Well, i'm a bit precious about this as many people poo-poo the desk without first listening to it. I've worked on some decent desks and I'd say that this console holds its own pretty well - 32/16/16, fully modular, +4db, Balanced everything, tt patchbay and VU meters, inserts everywhere, 4 band eq + 8 Aux sends all for less than US$1000 - a steal in my opinion. It smokes a Mackie 8 bus. I saw one go on eBay recently for $700.

My critisisms are:

- Sound a little 'smeared', particularly if you're heavy on the eq
- not as open as other name desks
- Mic pres are a not fantastic, which is why I don't use them

Oh yeah, it's a Tascam M600.

Thanks again for the help!

Chris

U101 - NJM2041SD:

Bipolar, Dual opamp. Low-noise version of the industry standard 4558.

Specs:
http://www.njr.co.jp/pdf/ae/ae04017.pdf

Could probably be subbed with very good results by a LM833.

Jakob E.

Why not bypass all the caps in the audiopath with a 0,1uF polyprop too, if there's space you can put them on the trackside!

You know I have to say that most people who claim to hear differences in capacitor brands after a recap in fact can't correctly identify them in a blindfold test. -There is a lot of "power of suggestion" at work there!

Leaving that potentially contentious subject alone for a bit, here's a few things to contemplate... SSL's cap failure rate slowed dramatically when they began to use two caps nose-to-nose and tie the junction to the positive rail through a high value resistance. -It completely prevents the cap being reverse-charged, and also eliminates the slightly nonlinear region around the positive/negative transisition. To this day, the SSLs built in this manner do not suffer the catastrophic failure rate that characterises Neve V-series consoles. -I'm not sure if I'm misunderstanding Jakob's use of the term non-polarised, but all of the SSL E/G sonsoles that I see have what I would describe as polarised caps in the signal path, and the vast majority are in pairs, with the junction charged from a rail through a high-value resistance. (9000s don't have caps in the signal path anywhere after the phantom-blockers, all the way to the console outputs...)

Me I'd steer very clear of the "double-ended" non-polarised caps... -they seem to bring all of the disadvantages of polarised caps, just for the marginal advantage of being theoretically capable of coping with a DC offset which could be slightly positive or negative...

Rupert of course nowadays keeps the signal away from the nonlinear 'transitional' area by biasing the op-amps output away from zero by a volt or so, which also keeps the cap predictably biased, but that's far too invasive ato contemplate as a retrofit mod!

-My take for what it's worth is that polarised caps used correctly should be capabale of outperforming any non-polarised variants. If you thing of the non-polarised variety as two caps back to back, and if you worry that inserting each cap can have negative effects, then think of using non-polarised caps as using twice as many...

I've used them in small speaker crossovers... It's one of the few times that I really have been able to hear a difference. -that was about 25 years ago, and I've never bought any since! -Of course, a speaker crossover is a very low-impedance, high-current test, but it shaped my opinions of them quite strongly.

Of course, let your ear be your guide. Don't let me or anyone else tell you what sounds good.. and try not to expect something to sound better because you think it should... this is tough to do, and I often fool myself the exact same way!

Let us know the outcome!

Keith

Originally posted by Jakob Erland:
U101 - NJM2041SD:

Bipolar, Dual opamp. Low-noise version of the industry standard 4558.

Specs:
http://www.njr.co.jp/pdf/ae/ae04017.pdf

Could probably be subbed with very good results by a LM833.

Jakob E.

This is why I love the internet, and in particular this forum. Thanks Jakob.

I just looked up the LM833, I don't think it is available in an inline configuration. Is this the weakest link? Do you think it would be worth the trouble? I know it's an open-ended question but if it is the weak point then I would definitely consider its upgrade.

Thanks again!

Cheers

Chris

Keith

I used the two caps to the + supply in a phone pre I built about 1981. This was after reading the Jung Marsh cap articles.

FWIW IMO the ISA110 might be better built with a single power supply so the electros would not be "rev voltaged" One might be able to use one of the BB opamps that could run at +-24 of +48 for both Phantom and Main power.

FWIW I think I can hear junk/bad electros: However once I replace them with ones that test good on a sencore lc102 I don't think I can tell well built caps apart of a good brand name. Most of the time the ESR going bad or bad seems to be what I can hear.

Yes, I've re-capped Studer A800s before now, and the difference is night & day... however in those cards the caps ry ouot like you wouldn't believe! -Many of the ones I pulled measured under 10% rated value. -What triggered the re-cap was when the LF EQ calibration adjustment wouldn't make it all the way up to flat...

Audible, measurable... improvable!

Good caps to good caps though... I can't tell the difference, a nor can most folks that I've run comparisons with...

Yeah, I have an LC-102 as well... it's a blessing when I've wound my own inductors! -realy easy to use!

Keith

Sorry Keith, just to catch up a minute, what did you mean when you say the caps are nose to nose on the SSL outs, did you mean two lytics with the positives facing out and the negatives tied with resistors to a posiive rail?

Hmm... that would smoke nicely!

Either tie the two positives together and link the junction to the positive rail through a high-value resistor, or tie the two negatives together and link the junction to the negative rail through a high-value resistor.

-Either way works, but if you tie them to the wrong rail, they won't last long!!!

Keith

I thought not! thanks for making it clear though.

Was this a Colin Sanders idea - he of soldering a stack of 741s together - or one of his minions?

I'm pretty sure that it was one of the design chaps there, although they almost certainly didn't invent the idea.

Colin was a truly great man. He built a better mousetrap and saw things through. -I don't know of anyone with a bad word for him. -Paralelling 741's... well, you try what you wanna try, -right!?!!!

Keith

Agreed.
I stayed with his brother in London once years ago (old family friend), he had a 6 storey town house in Vauxhall with about a dozen grandfather clocks (one on every stairway and every landing) and at least a hundred other clocks that he had rescued from scrap. Guess they had the same genes.

Funny thing was though, he had no idea whatsoever of the significance of SSL consoles; just one of his brother's projects that once occupied the family livingroom.

Tales abound of instances such as him offering to get out of bed in the middle of the night to drive people to their hotel when they finished their session in the wee hours... (at the in-house demo room)... definately a gent.

I briefly played the organ at a church back in England that had a Solid State Logic stop control system... From tiny acorns, Mighty oaks do grow!

Keith

Out of curiousity, what indications if any will you get that caps are deteriorating or in need of attention.

I have a Soundcraft Spirit Studio 24/8 that occasionally makes noises..........originating from the Master board and have been wondering if it is a cap problem.

:cool:

Hi String,

Symptoms of failing caps, in order of appearance:

Symptom #1: Scratchy pots
Symptom #2: Scratchy switches
Symptom #3: Loss of low-end
Symptom #4: Distorted low-end.

Intermittent problems are quite rare as a result of failing caps - unless we're talking PSU caps, which are a completely different story..

Jakob E.

Originally posted by Jakob Erland:
Hi String,

Symptoms of failing caps, in order of appearance:

Symptom #1: Scratchy pots
Symptom #2: Scratchy switches
Symptom #3: Loss of low-end
Symptom #4: Distorted low-end.

Intermittent problems are quite rare as a result of failing caps - unless we're talking PSU caps, which are a completely different story..

Jakob E.

Hi,

Gee, I thought the scratchy pots were dirt and dust!

FYI I've almost finished re-capping my modules, WIMA polyesters for C101/102 and Panasonics on the rest.

There's about 30 per channel in total (including the 'B' aux module) x 16, and boy it is a long process. Forget the 32 input channels...

Then I thought about those who have recapped the Neve VR series when the caps went bad... 200 or so per channel. Ouch.

Looking forward to my next DIY - the calrec EQ.

Cheers for all those who helped out!

Chris

Hi Chrisv

Have you find the schematic for your M600 ?
I have one also , that I like make some modification

the .JPG file is the picture of input module of M600 ?

it's long time a go , but if you can help me

Thank you very much
I am Canadian

Michel

You're asking question about folks from seven years ago. I'm not sure any of them have been seen around here in at least five.