Tantalums in the signal path?

[Sergimm]

I have a new-to-me Luxman 1030 with all original components that works well and sounds good so far. While looking at the circuitry to evaluate if there were any clear components that had to be changed (leaky capacitors, darkened resistors, blackened transistors, etc.), I found the two only lonely tantalum capacitors on the board, one for each channel. I attach both a picture of them and a picture of the schematic with them marked.

Now, I've read all the horror stories about tantalum, but I've also read that they make sense sometimes and that they shouldn't be blindly substituted by lytics or films. But I have some questions:

• Why would the designers use only two tantalums in the whole circuit for these particular positions? Definitely not for size, given that they are 3.3uF and are surrounded by 3.3uF electrolytics, roughly the same size.
• One thing I have read time and again is that if the tantalum are in the signal path, they should definitely be replaced if possible by film caps. I believe they are in the phono signal path, am I correct? If so, should I replace them?
• In case I replace them, what would be a proper replacement, keeping in mind their position, values, etc? Would a Wima MKS be good, or could I do better than that?

 

[jheu02]

One of the things tantalums are noted for is low leakage, which translates to noise in the audio path. Unfortunately, they can have a harsh sonic signature, which negates the good point of low leakage. So yes, replace those two either with film or Nichicon's UKL Series. Looks like there's space for Wima or other film cap with about a 5mm lead spacing. Some use bipolar caps in the audio path, so if you go that route, I'd use Nichicon UES "Muse" bipolar cap. Looking at the scat and photo, C219 and C220 (0.47uF/35V) should also be replaced by film caps. Not sure what the LR above the cap number in the scat means, but probably something to the effect of specialized cap (low noise, low leakage, etc) You'll need to look at 50V, 63V or 100V for the film cap replacements.

How are you faring over there COVID-wise? I always enjoyed my layovers in Barcelona when I got the chance to get there.

 

[Hyperion]

except that they last longer than electrolytics

Yep, and they are also very low leakage, I can't understand why we find them in the signal path so often. Their best use used to be for supply line decoupling in digital circuits, but even for that there are other alternatives now.

(Beat me to it John )

I would plonk a MKS2 there without batting an eyelid.

Me too.

 

[Okke 70s]

Well, that is an ever-lasting discussion you're getting in here. There are folks around here more knowledgeable about the matter than me, but let me try to give you some answers:

1: There could be a lot o reasons for any specific design choice, and they probably are related. In this case it would seem logical to look for a very low leakage/noise capacitor.

2. Because, yes, it is in the soundpath. And at the most "sensitive" point of the entire amp: as an input coupling capacitor before the phonestage. Any distortion coming from this capacitor is amplified, then amplified and then amplified again. If they work well replacing them is defendable but not strictly needed.

3. A MKS2 might do well. If could also sound a bit dry or hollow. I have quite divergent experiences with them in the soundpath myself. Even had amps running hot because of them. Yet other amps really gained in clarity. You could just try then. Alternatives might be fancy electrolitics like Nichicon KA series, mylard type capacitors, other metalfilms.. In my experience you just need to experiment to find out what sounds best in that particular place.

 

[llwhtt]

Very low leakage is the reason here, you don't want ANY DC voltage on the coils of the phono cartridge.

Craig

 

[Lavane]

I found a hand full in a Marantz preamp I had. I'm guessing maybe space was the problem due to the R&L channel boards being closely stacked on top of each other. Another was a Realistic integrated amp. The preamp board was very narrow with little space as well. With the Marantz I went with small Nichicon electronics. I was afraid film caps.l would have changed the sound too much. The Sony 3200F power amp I have had tantalum coupling caps. I did replace with films on that one.

 

[CoastinHome]

Why would the designers use only two tantalums in the whole circuit for these particular positions? Definitely not for size, given that they are 3.3uF and are surrounded by 3.3uF electrolytics, roughly the same size.

It’s probably just shy of 35 years ago that I asked the same question of Stan Warren (cofounder of PS Audio) as the Consumer Electronics Show in Las Vegas. Stan has a reputation for being a purist and minimalist who selects the components he uses through double-blind listening tests. I was a newly degreed EE at the time -- and was trying to learn from one of the gurus. We had met previously as PS Audio used to be in my college town of San Luis Obispo CA.

Stan and I were discussing the design aspects of his new Superphon DM-200 power amplifier (which I ended up purchasing from him at the show). The DM-200 was one of his products which used tantalum capacitors throughout the signal path. It and his "Revelation" preamps were very highly regarded "budget" preamps at the time (and may well still be).

Stan explained at the time that from his experimentation, solid dielectric capacitors (tantalum, silver mica) consistently provided better sonic performance when used in the signal path. I recall his comment that the capacitors in his designs remain powered up at all times and are biased in the middle of their operating voltage range. He found that keeping the capacitors charged provided both the best sound and greatest longevity. (I'm not sure if the front end of the power amplifier remains at idle -- probably not.)

Not sure what the current thinking is in this regard but those tantalum capacitors may have been very carefully considered by the designer. If you're changing out the capacitors you might want to carefully evaluate the before and after sonics.

 

[Blue Shadow]

What about changing the value of the input cap to a lower value? Maybe allowing a nice high grade audio cap to be used in that position? Since these are generally film caps which don't leak compared to electrolytics, most any of these audio caps could be used. Can this be done or does that cap need to be 3.3µF?

The 0.47 output cap from the phono stage could also be replaced with a nice film cap. Partsconnexion has a number of these in different brands. End up with a bespoke phono stage.

 

[roger2]

As has been said a couple times already, Wima is the obvious choice for those 3.3uF input caps. Plenty of room on the board and the lead spacing appears to be 5mm. And the Wima MKS2 is a better cap than any electrolytic including audio types or KL.

I agree that the .47uf output cap would benefit by upgrading to film as well. Wima would be a good choice for those too, and they could be ordered along with the 3.3s . Might want to verify lead spacing before ordering. 5mm lead spacing is what most Wima's are, but some are available in 2.5mm. And another option is the longer leads that could be bent in or out on the 5mm Wimas that are designated cut tape (CT) packaging.

 

[jheu02]

I pulled up the full SM and looked at the rest of the scat. LR stands for Low "Lack" according to the scat notes...poor translation from Japanese or a typo...Low Leakage. Elna LL caps were typically Orange sleeved. Looks like your dark blue ones might be Matsushitas. So, here's a couple more you might want to replace in the audio path...

Tuner Section
C127 (3.3uF/50V, nothing specified but is audio out of the HA-1137W chip Pin 6)
C130 (3.3uF/50V, same, input to HA-1156 chip Pin 2)
C155, 156 (are both Tant. I'd only replace IF you distrust tantalum failure modes, not audio path)
C159,160 (3.3uF/50V nothing special specified but use film or LL)
C164-167 (0.22uF/35V Tantalums)

Tone Board
C309,310 (1uF/25V LL)
C313, 314, 329, 330 (3.3uF/25V LL)
C323,324 (3.3uF/10V LL)

Main Amp
C401, 402 (3.3uF/10V LL)

You may be able to get a discount on the 3.3uF caps as I count 15 if you also replace C156. But I don't know whether you'll be ordering from an online vendor such as Mouser from the UK or going with a local shop.

 

[Sergimm]

I pulled up the full SM and looked at the rest of the scat. LR stands for Low "Lack" according to the scat notes...poor translation from Japanese or a typo...Low Leakage. Elna LL caps were typically Orange sleeved. Looks like your dark blue ones might be Matsushitas. So, here's a couple more you might want to replace in the audio path...

Thanks for that list! Just FYI, all the Electrolytics on the board are Marcon caps, not Matsushitas.

 

[Sergimm]

One of the things tantalums are noted for is low leakage, which translates to noise in the audio path. Unfortunately, they can have a harsh sonic signature, which negates the good point of low leakage. So yes, replace those two either with film or Nichicon's UKL Series. Looks like there's space for Wima or other film cap with about a 5mm lead spacing. Some use bipolar caps in the audio path, so if you go that route, I'd use Nichicon UES "Muse" bipolar cap. Looking at the scat and photo, C219 and C220 (0.47uF/35V) should also be replaced by film caps. Not sure what the LR above the cap number in the scat means, but probably something to the effect of specialized cap (low noise, low leakage, etc) You'll need to look at 50V, 63V or 100V for the film cap replacements.

Yeah, I'm not sure I have the patience to try many different types of caps and try to hear differences between them. I've heard that MKS2 and other films may be too "gritty" compared with audio-grade electrolytics, but I'm definitely bieased towards film ones bc of their other advantages and longevity. Btw, when researching low leakage caps, I found this scathing review for Wima films regarding leakage, which made me a bit less confident about using them.

How are you faring over there COVID-wise? I always enjoyed my layovers in Barcelona when I got the chance to get there.

Thanks for asking. We were doing great for the last couple of months, but it seems that number of infections are ramping up quickly now, just in time for the start of school year next Monday . Still, trying to keep up the optimism!

 

[Sergimm]

Well, that is an ever-lasting discussion you're getting in here. There are folks around here more knowledgeable about the matter than me, but let me try to give you some answers:

1: There could be a lot o reasons for any specific design choice, and they probably are related. In this case it would seem logical to look for a very low leakage/noise capacitor.

2. Because, yes, it is in the soundpath. And at the most "sensitive" point of the entire amp: as an input coupling capacitor before the phonestage. Any distortion coming from this capacitor is amplified, then amplified and then amplified again. If they work well replacing them is defendable but not strictly needed.

3. A MKS2 might do well. If could also sound a bit dry or hollow. I have quite divergent experiences with them in the soundpath myself. Even had amps running hot because of them. Yet other amps really gained in clarity. You could just try then. Alternatives might be fancy electrolitics like Nichicon KA series, mylard type capacitors, other metalfilms.. In my experience you just need to experiment to find out what sounds best in that particular place.

Thanks! Do you have any recommendations for other brands for films? I personally used KEMET in the past, ut I'd lie if I'd say I heard any difference.

 

[Sergimm]

What about changing the value of the input cap to a lower value? Maybe allowing a nice high grade audio cap to be used in that position? Since these are generally film caps which don't leak compared to electrolytics, most any of these audio caps could be used. Can this be done or does that cap need to be 3.3µF?

The 0.47 output cap from the phono stage could also be replaced with a nice film cap. Partsconnexion has a number of these in different brands. End up with a bespoke phono stage.

I'm not a fan of changing values, mainly because I'm no expert and I have no reason to doubt the engineers who designed the amplifier. But that's just me...I'm sure somebody with some real experience and knowledge could come up with interesting tweaks to the circuit, given the developments in the field these last decades.

 

[Sergimm]

It’s probably just shy of 35 years ago that I asked the same question of Stan Warren (cofounder of PS Audio) as the Consumer Electronics Show in Las Vegas. Stan has a reputation for being a purist and minimalist who selects the components he uses through double-blind listening tests. I was a newly degreed EE at the time -- and was trying to learn from one of the gurus. We had met previously as PS Audio used to be in my college town of San Luis Obispo CA.

Stan and I were discussing the design aspects of his new Superphon DM-200 power amplifier (which I ended up purchasing from him at the show). The DM-200 was one of his products which used tantalum capacitors throughout the signal path. It and his "Revelation" preamps were very highly regarded "budget" preamps at the time (and may well still be).

Stan explained at the time that from his experimentation, solid dielectric capacitors (tantalum, silver mica) consistently provided better sonic performance when used in the signal path. I recall his comment that the capacitors in his designs remain powered up at all times and are biased in the middle of their operating voltage range. He found that keeping the capacitors charged provided both the best sound and greatest longevity. (I'm not sure if the front end of the power amplifier remains at idle -- probably not.)

Not sure what the current thinking is in this regard but those tantalum capacitors may have been very carefully considered by the designer. If you're changing out the capacitors you might want to carefully evaluate the before and after sonics.

That's a great story. Yeah, I'm still not sure of what caps I will end up using there. I would even be open to use modern tantalums, because if they are still made and used nowadays given the great progress on film and electrolytic caps, it must mean that they still have a strong use case for modern circuits...and they would probably provide the closest "sound" to the one the original engineers wanted in that position.

 

[Okke 70s]

Thanks! Do you have any recommendations for other brands for films?

If you are happy to spend a bit more, orange drops might be the thing. But I doubt it would sound different from the wima's either, in your actual setup.

It only 2 for now, so just throw in the wima's I'd say.

 

[ConradH]

IMO, the big concern with phono stages is isolating all possible DC from the cartridge. Tantalum generally has very low leakage, but it's a lot less predictable with age. I've seen a lot of tants develop leakage with age. AFAIK, higher value films were more expensive back when the gear was built, so tantalum was probably their better choice at the time. Today I'd go with film. FWIW, if the value of a coupling cap is sufficiently large that no audio voltage appears across it in operation (and that's how coupling caps should be chosen), it shouldn't affect the sound. It can't- no significant voltage appears across it. If you compare the impedance of the cap to the impedance of the cartridge, the ratio is large, so again, it can't have much affect on the sound. That's for caps that isolate the cartridge. For other caps, you have to look at the circuit.

 

[roger2]

......

3. A MKS2 might do well. If could also sound a bit dry or hollow. I have quite divergent experiences with them in the soundpath myself. Even had amps running hot because of them. Yet other amps really gained in clarity. You could just try then. Alternatives might be fancy electrolitics like Nichicon KA series, mylard type capacitors, other metalfilms.. In my experience you just need to experiment to find out what sounds best in that particular place.

Hi Okke,

I am curious about your comment of having amps run hot because of Wima caps. I am having trouble conceiving of a scenario where that would be possible. Could you please elaborate on your experiences where Wima caps caused your amps to run hot?

 

[Okke 70s]

Hi Okke,

I am curious about your comment of having amps run hot because of Wima caps. I am having trouble conceiving of a scenario where that would be possible. Could you please elaborate on your experiences where Wima caps caused your amps to run hot?

Well, I'm not quite sure about it all, because it was years ago..

But it probably was with the JVC A-X4 (love that amp!). After recap, the sound was off: very thin, his and sharp. There was also a soft hissing sort of distortion. Maybe some kind of resonance? It did run the amp hot anyway.

After some fuzzing around I found that replacing the wima's with electrolitics resolved the problem. For each channel separately. Double checked on capacitance etc, but was never really able to pin down the couse.

They were most likely the 100v version, but that should not be an issue right? Another difference is that the wima's are bipolair.

Well, I was just happy enough that the amp was working well again.

 

[Thorsten R.]

That is an interesting discussion! I have to admit that I definitely do not like tantaliums. That has a reason: When I repaired HiFi equipment for my living I had to work on a lot of open-reel tape recorders. Especially Uher and very often the Report - series (I still love that things!). The absolute most frequent failure was caused by diverse Tantaliums. They failed at a lot of other equipment too. I have to admit CoastinHome´s story has a lot for it. However just several days before I have recapped a small Rotel RX-303 and used WIMA MKP2´s as signal coupling caps where possible. Especially in the phono stage and between the tone control and the input of the differential amp. This breathed a lot of life into the sound of this small, relatively cheap receiver! Especially when I listen with my trusted Sennheiser HD 250 Linear. So I second film caps, especially Wima MKP2.