Best capacitors to recap Roland Jupiter 8 synthesizer

[rokuez]

maybe ask a mod to move this into geekslutz u might get better respnoses

[Reptil]

nah, the information is all right here.
do a search next time ok?
thanks

[scottrod]

Originally Posted by Reptil

do a search next time ok?

Not to be contrary, but in the third sentence of my original post I indicated I had read that Jupiter 8 post, but my questions 2, 3, and 4 were not answered very well (or at all) so I thought I would simply ask for a bit more discussion. Perhaps I don't understand the intent of GS forums?

My question number 1 was discussed breifly by Don Solaris, but I wanted a few more people chiming in if that's not being too repetitive.

Would just like to get a few final issues tied down before i launch into a project i might regret, and thought I might get a few thoughtful responses here. Should I just have revived the old thread?

[Don Solaris]

Originally Posted by scottrod

1. I know it's good for the synth's longevity in general, but do you guys think this is really going to make a sonic difference? I'm skeptical, but will take the advice of the more experienced.

Best capacitors to recap Roland Jupiter 8 synthesizer

2. Do I need to replace ALL the caps, or just those on the audio path?

Apparently ALL. Why would one replace just capacitors in the "audio path" ??

3. Replace all decoupling caps on every IC power pin? Just the electrolytics, that is, are ceramic coated caps also prone to changes over time?

Best capacitors to recap Roland Jupiter 8 synthesizer

4. The T8 uses the pots and pressure sensors as CV to the CEM filter processor; no audio goes through them, but they do have a decoupling cap as well. Should these caps be replaced?

Apparently, ALL.

[Reptil]

Originally Posted by scottrod

Not to be contrary, but in the third sentence of my original post I indicated I had read that Jupiter 8 post, but my questions 2, 3, and 4 were not answered very well (or at all) so I thought I would simply ask for a bit more discussion. Perhaps I don't understand the intent of GS forums?

My question number 1 was discussed breifly by Don Solaris, but I wanted a few more people chiming in if that's not being too repetitive.

Would just like to get a few final issues tied down before i launch into a project i might regret, and thought I might get a few thoughtful responses here. Should I just have revived the old thread?

maybe. the reason behind my question is that any forum will become unreadable and unsearchable if all discussions and information about the same subject are scattered throughout many threads.
maybe a morph is in order. maybe not.

Capacitors are funny things. Yeah what he ^^ said, change all electrolytics in one go.

[scottrod]

Sorry to waste everyone's time. Please delete the thread.

[Muser]

All I know about it is that, I have talked to the UK Roland guy who they used to fix Roland equipment. He put Audio grade capacitors on the output stages of Duran Duran's JP8 and Nick Rhodes asked Him how come it sounded better.

He himself attributes it to those Audio Grade output caps.

don't ask me though.. because I can't confirm anything but He said that to me.

[Don Solaris]

Originally Posted by Muser

He put Audio grade capacitors on the output stages of Duran Duran's JP8 and Nick Rhodes asked Him how come it sounded better.

OK what do we do next?

[Muser]

Originally Posted by Don Solaris

OK what do we do next?

plug some leads in

no seriously though. I'm just addressing one specific thing.

all the other aspects you are discussing here I'm not specifically addressing.

That's why I started my sentence with (The only thing I know).

[Don Solaris]

Originally Posted by Muser

no seriously though. I'm just addressing one specific thing.

Yeah, but didn't you noticed something strange in that story?

Like the fact he concluded it sounded "better".

- "better" what?



The fact he said "better" is more likely to please his tech, cause Nick is a nice guy.

Or his unit had some dry or leaked out capacitors (which is normal after 25 years of use) and that his Jupiter started to sound shitty due to increased high pass filtering. In such case, new capacitors would indeed improve the sound.

However, it is totally irrelevant are those labeled as "audio grade" or not. Because every capacitor, in low currents, low amplification, low frequency (such as a synthesizer) is already "audio grade" (!). In 0-22kHz there is literally no difference in the sound of capacitor A vs. capacitor B when placed in audio path of a synth.

I tried a lot of caps. Not by "it sounds better now" method but by actual null tests. Found no difference at all in the sound, just in quality (lifetime) and true values vs. declared value.


My bet: Nick just wanted to say nice word to his tech. thumbsup

[Muser]

Originally Posted by Don Solaris

Yeah, but didn't you noticed something strange in that story?

Like the fact he concluded it sounded "better".

- "better" what?



The fact he said "better" is more likely to please his tech, cause Nick is a nice guy.

Or his unit had some dry or leaked out capacitors (which is normal after 25 years of use) and that his Jupiter started to sound shitty due to increased high pass filtering. In such case, new capacitors would indeed improve the sound.

However, it is totally irrelevant are those labeled as "audio grade" or not. Because every capacitor, in low currents, low amplification, low frequency (such as a synthesizer) is already "audio grade" (!). In 0-22kHz there is literally no difference in the sound of capacitor A vs. capacitor B when placed in audio path of a synth.

I tried a lot of caps. Not by "it sounds better now" method but by actual null tests. Found no difference at all in the sound, just in quality (lifetime) and true values vs. declared value.


My bet: Nick just wanted to say nice word to his tech. thumbsup

I think the situation was that He has it checked out before tours, even if nothings wrong with it. This time He changed the caps on the output stage to the Audio grade ones.

I can only assume he had some basis for his own thoughts and how it corresponded with what His client said there.

not sure how you are doing a null test there. you mean by opening the filter on the synth and then running a before and after null ?

or do you mean by running a signal generator through the output board ?

I have some doubts about nulls to be honest. one doubt is the difficulty of setting a foolproof condition for what you are testing. also, that a single condition is just that, a single condition from which a general assumption is drawn.

but that's a different story.

[Yoozer]

This is the third time that this kind of thing happens - you blurt out a few sentences on a topic, someone else calls you out on it, and then you backpedal and say you don't know enough to discuss it on such a technical level, but....

You doubt null tests? What the...?

Do you know how these actually work?

You record signal one, you record signal two, you flip signal two, sum it with signal one, and if what's left is noise at an extremely low level, you've been imagining differences. It's thanks to this hypothesis that balanced cables even work, or certain topologies of highpass filters where the lowpassed signal is removed from the original.

And you're going to tell, of all people - Don Solaris - that he might've not set up things correctly?

Sorry for being a bit of a pain in the rear here.

[Xero]

this is just why you can't really rely on your ears for comparing ANYTHING in any sort of useful way.

we just don't hear well enough and are too easily influenced by outside factors

it's actually somewhat involved to do proper double blind testing, and as such, more often than not individuals fail to really get it right...and famous individuals are no more exempt from this...though i have a feeling it was more of a compliment to his tech, like some of the other people were saying... "yeah man, sounds great!! better than ever"

the scientific method involves defining a provable and repeatable thing, the human brain, not so much...

with the milspec capacitors in these things, i question why he needed to replace them at all, well other than the power supply ones anyway. Then again i wouldn't be surprised if his jp8 is beaten and battered from touring and what not...who knows though.....

[Muser]

Originally Posted by Yoozer

This is the third time that this kind of thing happens - you blurt out a few sentences on a topic, someone else calls you out on it, and then you backpedal and say you don't know enough to discuss it on such a technical level, but....

You doubt null tests? What the...?

Do you know how these actually work?

You record signal one, you record signal two, you flip signal two, sum it with signal one, and if what's left is noise at an extremely low level, you've been imagining differences. It's thanks to this hypothesis that balanced cables even work, or certain topologies of highpass filters where the lowpassed signal is removed from the original.

And you're going to tell, of all people - Don Solaris - that he might've not set up things correctly?

I didn't say Don had set anything up incorrectly.
I said what I meant and I even qualified the statement about the capacitors by saying at the very header of the sentence that it was the only thing I know about capacitors in relation to JP8.

yes I do know how nulls work. I spent a whole week setting up a nulling condition EQ where I was nulling a white noise & a swept sine signal against another white noise & swept sine. one of them had passed through a Roland pre-emphasis curve. I tried multiple EQ types and approaches to do this and I was observing very odd signal artifacts down around -90db to -110db on very low value EQ alterations 0.1dbs etc.

so I was nulling out an intentionally induced pre-emphasis.

when I say a week, I mean 7 14 hour days.

That experience left me with a lot of questions about what may be possible limitations of null tests.

There are more people who question the total veracity of what a null might tell you and under what conditions than just, Myself.

[Don Solaris]

Originally Posted by Muser

not sure how you are doing a null test there. you mean by opening the filter on the synth and then running a before and after null ?

A patch memory. It will load exactly the same settings for every parameter.

Originally Posted by Muser

or do you mean by running a signal generator through the output board ?

No need. Just record 10 seconds of saved patch on exactly the same level. I used Juno 60 (on purpose, because it is a DCO synth which provides stable frequency response, else you can forget a null test) and one single waveform. Then align their phases, invert, and you're done.

But much more fun are the actual patches with filter sweeps etc and listening to your ears without knowing which is which. I used cheap N.A. capacitors vs Panasonic and Elnas. No difference in sound at all. Since that day, when ever i hear "Bob put audio grade and wow sounds better" my stomach wants to throw away its content.

It really starts to get on my nerves when some "Bob" tells me it "sounds better". WTF? Does he got a hard disk recorder in his head? How the hell does he know it sounds "better"? I would understand doing A vs B tests of the actual recording, prior and after the recap. That would make sense.



As of Nick, his comments more sounded like "thanks". If those capacitors changed something in the sound, i'm sure he would address it by actually pointing at it with something like: "hey! i get much better bass response!" or "the high end is more clear now!". He would definitely not say "sounds better", because, face it, that doesn't mean anything.

[Muser]

Thanks for the response Don.

I understand why you might be very skeptical after having tried out the theory under a listening condition and a technical one.
I have a tendency to continually doubt my conclusions. I still doubt what I was seeing with the pre-emphasis EQ match.
I've viewed threads where multiple high end ADs are being tested and people are picking out different converters there. even while listening to the results through their (own) converters, which is a potential problem in itself. I would expect that those signals will null too. so I am just left with even more questions on that one.

btw, I sometimes treat GS as a database rather than a social or technical help forum. This is why I may seem to interject I think. I need to think up some kind of qualification so people can just hop over what I say, or ignore it, or just state their disagreements etc and then carry on.

I try to account for it but It's actually more difficult than it first looks.

I wouldn't mind hearing what you think about possible impacts of chemical capacitors on technology at some time.

[Don Solaris]

Originally Posted by Muser

I wouldn't mind hearing what you think about possible impacts of chemical capacitors on technology at some time.

There you go: Electrolytic capacitors and why to change them

thumbsup

[Catabolic]

FWIW the Panasonic caps get some mixed reviews over at Prodigy Pro forums for the DIY 1176 compressor. The opinion that they sound harsh or hyped is what I recall. This is from people who have built two units but have used different caps in the two. Full disclosure - haven't done this myself.

[Reptil]

a makeup amp in such a compressor would certainly reveal the character of a cap. I wouldn't say Pana's sound sterile, or hyped. Just clean. But that's my personal opinion and differences are very small between these caps. in synths even smaller
If you want a bit of color you can try BC, Vishay, Elna... often 85 degree caps alas.

[backindauk]

To all Jupiter 8 owners and techs alike; i just finished a parts spreadsheet for a complete recap.
It lists Digikey part numbers and quantities for all electrolytic capacitors. I am working on adding the exact board locations of each cap in the next couple of weeks. That task will be easier once i have a Jupiter 8 in for repair.

My list uses Elna caps for the boards as well as Panasonic & Nichicon caps in the power supply. (6800uf 35V Panasonic with the right lead configuration is not available - at least not in small quantities for a reasonable price)


If anyone wants the list, please email me. A couple of members already have it.

[gilwe]

I was asking myself the same question (which caps are the best for vintage synths recapping) and got to this great thread. Reading it all, I really wonder how mrhuge's JP8 sounds once the recap has completed... I'm considering getting a few of my synths recapped soon and would like to use the best once I do it. This the time you really start to see them leaking (green goof at the bottom of caps). Need to recap before the situation gets worse

What makes me curious about this very discussion is that when I fully recapped both a Korg MS-10 and a Yamaha CS-15 (a few years ago) I really used only what I had which were cheap chinease Jamicons (or any other cheap brand). With all that, I can't explain how good they sound now. Clean as much as any other new digital keyboard but still analog, fat and very very accurate. With all that, when getting to recap a polysynth, I'd really much like to use the best.

I wonder if mrhuge's selection of caps (all types of Nichicons) for his JP8 was a good choice. I have some pictures from an inside of a JP8 and there are many Sanyo caps in there. It still sounds awesome

[oldgearguy]

Originally Posted by gilwe

I was asking myself the same question (which caps are the best for vintage synths recapping) and got to this great thread. Reading it all, I really wonder how mrhuge's JP8 sounds once the recap has completed... I'm considering getting a few of my synths recapped soon and would like to use the best once I do it. This the time you really start to see them leaking (green goo at the bottom of caps). Need to recap before the situation gets worse

FWIW - most of the time that junk you see on the bottom of large caps is simply an adhesive the manufacturer used to help keep the caps in place. Back in the day, manufacturers thought about things like shipping, touring, dropping, etc and tried to make sure the gear could handle it. The big power supply caps are large and heavy and if they only relied on the leads to hold them in place, over time you'd have dry/cracked solder joints where you least want it.

You will also often see the axial caps held down with zip ties that are run through pre-made holes in the circuit board.

Bad caps (other than measuring them) can sometimes be determined by bulging tops. Radial caps were never manufactured with a convex top. If a cap is really leaking electrolyte paste, it's probably bad enough to severely affect the circuit it is in.

[dougt]

IMHO the only caps you might want to replace is the big electros in the power supply. The stress to the boards, components, circuit traces and pads recapping everything is not worth it...

[EofN]

Originally Posted by dougt

IMHO the only caps you might want to replace is the big electros in the power supply. The stress to the boards, components, circuit traces and pads recapping everything is not worth it...

Yes!

I can only imagine how much valuable, irreplaceable vintage gear is going to get totally trashed by a thousand well intentioned amateurs and their soldering irons. Them and UPS.

[gilwe]

hmm... well the "junk" I saw is greenish acid. Actually in the specific case it was a Jupiter and the leaking caps were all of the same type (if I'm not wrong they were all 47uf), so I guess such cases do need cap replacement to be taken. I have also recapped a Hammond B100 a few years ago of which just about all caps have already started to leak, just the same way.

[audioconsult]

Originally Posted by Don Solaris

Yeah, but didn't you noticed something strange in that story?

Like the fact he concluded it sounded "better".

- "better" what?



The fact he said "better" is more likely to please his tech, cause Nick is a nice guy.

Or his unit had some dry or leaked out capacitors (which is normal after 25 years of use) and that his Jupiter started to sound shitty due to increased high pass filtering. In such case, new capacitors would indeed improve the sound.

However, it is totally irrelevant are those labeled as "audio grade" or not. Because every capacitor, in low currents, low amplification, low frequency (such as a synthesizer) is already "audio grade" (!). In 0-22kHz there is literally no difference in the sound of capacitor A vs. capacitor B when placed in audio path of a synth.

I tried a lot of caps. Not by "it sounds better now" method but by actual null tests. Found no difference at all in the sound, just in quality (lifetime) and true values vs. declared value.


My bet: Nick just wanted to say nice word to his tech. thumbsup

but you know what they say about null tests? On some machione roland themself uses bipolar audio caps.. why would they do that when thee wouldnt be a difference?

[Reptil]

Originally Posted by dougt

IMHO the only caps you might want to replace is the big electros in the power supply. The stress to the boards, components, circuit traces and pads recapping everything is not worth it...

If caps lose capacitance, if their ESR builds up, or even start to leak, they need to be replaced, because it will interfere with the functionality of the machine.
Electrolytic caps have a limited lifespan. especially the old stuff from sixties and seventies, and cheap chinese caps that were popular a decade ago. If one goes, then the other electrolytics are logically not far behind. Some old highend caps like ROE keep soldiering on after more than 30 years. But that's not the quality of parts, what you'd usually find inside a synth. There are new capacitors now on the market that are very precise and don't mess up the electrical signal, and are very long life. The impact on the sound depends entirely on where these caps are in the circuit, and what function they have.
If someone doesn't know how to recap a synth, or mixing board, they should take it to a technician that can do the job, and has invested in the equipment and skills.

[gilwe]

Evey vintage synth made between late 60s and mid 80s will need a full recap sooner or later. It's a tedious job but except for it being time consuming, it is quite a simple task to do, and the only must-do task in order to keep a vintage synthesizer alive. Leaking caps will surely cause damage to the boards and nearby components. Caps which changes values could in a way cause ICs to go bad. They are cheap components... so, as long as you put the time to do that - you (or your synth) are more or less safe

The big question remains - which caps are the best to use ?... Don Solaris has recommended the Nichicons... I wonder what kind of feedback we will be able to get from the thread starter, a few years after his JP8 recap has been done (or hasn't?)

[Reptil]

Nichicons are good. So are BC (Vishay), Panasonic FC, FM and a couple others. Cornell Dublier is good. Older ROE are still measuring fine. All low ESR, stable capacitance. You can't see into the future but if a cap is specified for 5000 hrs at 105 degrees celcius, then logically it is to be expected to last a long time (longer than the original low specified capacitors) in a circuit that doesn't get hotter than 85 degrees.
I put some in an equaliser in 2010 and they still measure fine. Also have equipment that was recapped in the last decade, and those caps are also still good.
I wouldn't worry about it too much, just use good parts and don't fry the traces on the boards.

[Don Solaris]

Total recapping is necessary in some cases...

My Boss DE-200 unit had a strange issue where the sound of the delayed signal would be incredibly low in volume for the first 30 minutes until the machine would warm up. I couldn't figure out the problem. So i decided to check capacitors.

Turned out some capacitors were at 70% of declared value. So i replaced 30+ caps in the unit. Guess what. Delay now works perfectly from the first second! No more need for wait to warm up. (i used low ESR caps for replacement)

- Case Closed -