Good tantal audio caps? (Studer 089 console refurbushing)

Hi,
which good tantal audio caps would you recommend me?
I have to refurbish entirely a Studer 089
I checked Vishay 199D caps but the price is prohibitive, more than 4€ for one cap!
Just for one value it would cost me nearly 400€...
Many thanks

thekid777 said:

Hi,
which good tantal audio caps would you recommend me?
I have to refurbish entirely a Studer 089
I checked Vishay 199D caps but the price is prohibitive, more than 4€ for one cap!
Just for one value it would cost me nearly 400€...
Many thanks

Click to expand...

Do you know Al electrolytics made today are as good as Ta made 20 years ago? And much more reliable...

A properly applied solid tantalum cap can last nearly forever. There is no real wearout mechanism, only failure mechanisms caused by excessive voltage or current. Since most Ta caps used inside of a console see a controlled driving and load impedance, as well as fixed power supply voltages, a great many of them can last forever.

The only caps that could be at risk are those used to couple to an external connector, or those used to shunt a power supply rail that does not have properly constrained currents or voltages.

So, re-think the need to replace all of these caps. They are probably part of the sound of that console anyway, and if they're not broken, you should leave them alone.

Based on my one experience with a Studer 089 tracking drums it had a very distinctive and colored character, if I was replacing any parts I would endeavor to replace like with like, especially Tants with their known distortion characteristics (discussed several times over the years for Neve refurbs).  Electrolytics may be higher performance but in this case I'm not sure that's what I'd want.

Based on just one experience and with the greatest of respect for Abbey's input.  The 089 was chosen over a very nice Neve for the drum tracking and was anything but straight wire with gain.

thekid777 said:

Hi,
which good tantal audio caps would you recommend me?

Click to expand...

Like everyone said, don't replace the tantalum capacitors, there's no need or reason for it if they are working

replace all electrolytic caps if you want
use Panasonic capacitors you can use any series rated at 105 degrees

Monte McGuire said:

A properly applied solid tantalum cap can last nearly forever. There is no real wearout mechanism, only failure mechanisms caused by excessive voltage or current. Since most Ta caps used inside of a console see a controlled driving and load impedance, as well as fixed power supply voltages, a great many of them can last forever.

Click to expand...

That´s the theory. Real life is different. E.g. I´m currently refurbishing a Sontec 250c which has a lot of tants on board. 80% of them were in a state of failure. They are resistive down to a value of the surrounding support circuitry (DC servo), which causes serious misbehaviour of the circuit. And that´s only one example. I´ve seen this many many times. IME tants from the 90s on were a lot more reliable than earlier ones. In case of the Studer desk mentioned above I´d replace them all. If OP wants to stick with the original sonic signature then replace them with tantalums, if a cleaner sound is desired then use electrolytics, eventually with bypass caps.

The best solid Ta caps I've seen are the Kemet parts, but others seem to work as well. Yes, not all vendors' parts are the same, but misapplication (which usually means insufficient voltage de-rating) will cause perfectly good parts to fail. While I don't know about the Studer 089 in particular, Studer seems to understand 'the rules' and the original poster might have wanted to 'shotgun' the device because of the popular idea that 'electrolytic caps need to be replaced', and that's not the case with a solid Ta cap, properly applied. There's no liquid electrolyte to evaporate, etc. so the same wear-out mechanism that wet Al caps have is not present.

Once a Ta cap has settled in, its reliability increases, as small, current limited faults get 'healed' and eventually disappear once all voltage excursions applied to the cap have been applied. After that, there are no further over-voltage events that cause a localized short and re-forming of the Ta oxide layer, so the part truly does become more reliable.

Still, uncontrolled currents can cause additional failures, but for a cap inside of a circuit where the source and load impedances are fixed, and the voltages are limited, these failures will diminish over time, and thus there's no wear-out mechanism as such.

I would suspect that the Sontec circuits misapplied solid Ta caps, and that a replacement with higher voltage rated (or at least larger slug parts) would work to prevent future failures.

I recently refurbished an Eventide H3000 that had a lot of Ta cap failures, so yes, it can happen. In this case, it was from a regulator failure that applied an over-voltage to all of the supply bypass caps, and yes, a total shotgun replacement was required. However, I was able to use higher voltage rated caps (35V vs. 25V) for its 15V rails, so I'm certain that this failure won't happen again.

At the same time, I worry that if the regulator fails again, the Ta caps will not short and save all of the silicon by shorting the rails! At the end of the day, I think it's still better to prevent any and all components from failing, and not leaving vulnerable Ta caps there as some sort of supply rail 'crowbar' - that seems shoddy! ;-)

We're in agreement that solid Ta caps can fail, and we've both seen it. However, I question whether the caps in this Studer 089 actually failed, or whether it is just a well intentioned 'shotgun' because of the pervasive idea that 'all electrolytics need to be replaced over time'.

The first thing in any capreplacement program is to remove a couple of caps and measure them. Are they at or above their original value? Then they probably do not need to be replaced. There are so many people who have this idea that all caps should be replaced. They are wrong. Why change them if you dont need to? I recently measured a Neve BCM10. 7 of 10 channels measured firne for frequency response. The other three channels showed response variations indicative of cap failures. One channel in particular had frequency response 5 dB away from flat. Funnily, none of the console users had noticed this.

Gary's tant replacement rules...

Replace ALL Tants across power rails unless you like fires. :'(

Replace the series coupling cap  C32 on Ampex record
cards used in AG440 and MM series machines unless you
like fried record heads. :'(

GARY

gar381 said:

Gary's tant replacement rules...

Replace ALL Tants across power rails unless you like fires. :'(

Click to expand...

I'll disagree with this, but only for specific situations. The Eventide H3000 that I recently rebuilt was properly designed with Ta power supply bypasses because the box ran _so_ hot that there was no hope that any wet Al electrolytic, even modern ones, would survive for more than a few years. They used quality 25V solid Ta caps for a 15V rail, and that would have worked with essentially no safety margin, except when the linear regulator failed and dumped 20V onto the +15rail, causing the 25V Ta caps to short.

The simple thing to do is to replace the 25V Ta caps with 35V parts, which will probably withstand 20V, and will most certainly withstand 15V.

The basic rule with solid Ta caps is to 'de-rate' them appropriately. If you want a cap to work with let's say a 15V rail, you need to use 2x that voltage or more. So, there aren't 30V Ta caps, but they do make 35V caps, so those are good. 25V caps will probably work, but they're at the ragged edge already, so if you want few failures, 25V caps are not a good idea for 15V rails. 2x de-rating is a good rule, and it will prevent many fires.

Current spikes will also kill solid Ta caps, so if there is any chance of uncontrollably large currents through a bypass or coupling cap, one should not use a Ta cap. Caps that drive external connectors are also susceptible to this - if that jack were shorted or subjected to odd external voltages, this may cause the Ta cap to fail.

These applications are very troublesome to me, and I agree with you that if one wants to avoid fires, do not use solid Ta caps where they could be subject to  external voltages or impedances that are not well controlled. Inside of a circuit, they can work well, but this is not the case if they are exposed to external nodes.

Ampex chose tant rail caps in an appropriate  manor in this ATR100. ???




A very expensive ATR100 fix  :'(

GARY

Yeah, that went as far from 'well' as possible! I hope that failure didn't cause a structure fire...!

Personally, I will never design Ta into a device, unless it's for some very specialized place, like a time constant cap that is completely power and voltage limited, e.g. the CV cap in an 1176 style detector. Still, we sometimes are asked to revive old boxes whose character depends on this stuff, so figuring out a way to get it right, despite what the original designer did, is sometimes important.

Ta caps have been force-fed on designers in the 70's and 80's because they were a BIG money maker. They were pared with all virtues, and the drawbacks thoroughly ignored. Today, they are well-known, and I can't think of any circumstance where I could justify the choice of Ta in an audio application.
Some of the Ta cap manufacturer recommendations are ludicrous:

"Solid tantalum capacitors with crystallization are most likely to fail at power-on... To prevent such sudden failures, manufacturers recommend:
50% application voltage derating against rated voltage
using a series resistance of 3 Ω/V "

I believe they should be avoided, because of the potential risks and because Al caps and other technologies have made huge progress.