The basic parts should be good.
What I would look at is where they are in circuit, the operating voltage and what those are rated for. Just because it is a new build and design doesn't mean they were used properly. Maybe just old habits from service work, but I still put much higher voltage rated solid tantalum capacitors in than you would for an electrolytic. They will not tolerate voltage spikes or reverse polarity.
What I would look at is where they are in circuit, the operating voltage and what those are rated for. Just because it is a new build and design doesn't mean they were used properly. Maybe just old habits from service work, but I still put much higher voltage rated solid tantalum capacitors in than you would for an electrolytic. They will not tolerate voltage spikes or reverse polarity.
Checking is normal and recommended. One would assume a manufacturer knew in 2020 how to implement tantalum caps but checking is better.
Attachments
Exactly as described. Check to verify. Usually there is a brand symbol, a value and voltage and a datecode. Still logic says these are not the famous bad tantalum caps of decades ago. These have a reputation so bad that current tantalum caps still suffer and about every designer is familiar with the matter.
I know you think I am retarded but the point is I think it is completely logical to check such trivial things when looking at a board and the parts.
The bag you show has no labels. I would discard those if there can not be determined what brand/series and age they have.
I know you think I am retarded but the point is I think it is completely logical to check such trivial things when looking at a board and the parts.
The bag you show has no labels. I would discard those if there can not be determined what brand/series and age they have.
Last edited:
After the UKZ, and UKG, Nichicon seems to give up the very low leakage UKL serie due to the low demands (Mouser notification).
It looks like all the Nichicon that say "audio" on the data sheet are being discontinued...
Regardless, they still make good capacitors.
Regardless, they still make good capacitors.
Yes, indeed.
There are good substitutes for UKL with other brands. Panasonic makes low leakage and low ESR caps, as does Wurth. Neither are marketed as audio caps, but they perform very well, as reflected in their specs, but also in my testing with an HP 4274A and a much newer TH2832.
Me who thougth the UKL was the lowest tht lythic leakage cap with maybe one Rubycon on the specs...
Interrested by the ref you have. Can be handy.
Interrested by the ref you have. Can be handy.
Soon ill be replacing electrolytic caps on the boards of a Philips CD303 I recently picked up. It probably recently had its 40th birthday - hope the previous owner organized a nice getogether to celebrate. At least they didn't mess with it, as it looks almost factory original. Great sound too. It sounds like I'll be retiring my TDA1541A-based DAC and go back to spinning discs.
Choosing caps for obsolete series of 'lytic caps is a bit tricky given the ridiculous of number of products available. I try to stay as close as possible on all specs, from physical dimensions and form factor to electrical specs down to esr. For example, it's nice that the many 22uF and 33uF axial ones appear to have replacements in current production Vishay BC components.
I wonder about 'low leakage' specs tho. For some 2200uF and 1500uF smoothing caps on the power supply board (right after the diode bridges), the caps that match the original most closely (except for the leakage current spec, as the originals were quite leaky according to an old Philips data book) are Nichicon UKLs. Is there any reason not to use UKLs there?
I always considered low leakage electrolytic caps as something exotic and use them where explicitly called for (like in the feedback network of an opamp based discrete series voltage regulator that I use a lot - I will be stocking up as they're being phased out).
But shouldn't every cap be 'low leakage' in principle? Who would want a leaky cap anywhere? What's a cap that's not 'low leakage' anyway? Pretty leaky, medium leaky, super leaky? I reckon that most of the time marketing departments don't ask themselves that question.
Choosing caps for obsolete series of 'lytic caps is a bit tricky given the ridiculous of number of products available. I try to stay as close as possible on all specs, from physical dimensions and form factor to electrical specs down to esr. For example, it's nice that the many 22uF and 33uF axial ones appear to have replacements in current production Vishay BC components.
I wonder about 'low leakage' specs tho. For some 2200uF and 1500uF smoothing caps on the power supply board (right after the diode bridges), the caps that match the original most closely (except for the leakage current spec, as the originals were quite leaky according to an old Philips data book) are Nichicon UKLs. Is there any reason not to use UKLs there?
I always considered low leakage electrolytic caps as something exotic and use them where explicitly called for (like in the feedback network of an opamp based discrete series voltage regulator that I use a lot - I will be stocking up as they're being phased out).
But shouldn't every cap be 'low leakage' in principle? Who would want a leaky cap anywhere? What's a cap that's not 'low leakage' anyway? Pretty leaky, medium leaky, super leaky? I reckon that most of the time marketing departments don't ask themselves that question.
Hi kunlun121,
Honestly, Electrolytic capacitors are non-critical despite what people say. Each cap varies widely in the same batch for example. ESR may matter for power supply filters, but if this is critical they would use high frequency types. Not common when this was built.
My strong advice? Buy good quality capacitors from authorized distributors. The larger case will be a better part, don't put one in too large. Higher voltage ratings are generally better and help to have the same size case. These capacitors will not change the sound. Others will, but not many here talk about what actually does matter.
I've been a paid technician for almost 50 years, servicing longer. Studied parts since I began, and I have the equipment to do it. Serviced real high end audio and recording studio equipment. I'm just trying to save you effort and money while suggesting the best outcome for you. What you do is totally up to you. For many, parts are a religion based entirely on beliefs and anecdotal reports without real testing. "White papers" are just advertising copy with some technical stuff thrown in that often isn't applicable or truthful (it's an ad, remember?).
I'll say it one more time. There is no best make / model of capacitor. Each part has characteristics that make it better for one circuit location over others, and production tolerance makes the difference between good brands and models insignificant. ESR has no bearing in coupling applications, except in series with a load (higher current). What differences matter even at one ohm when the circuit impedance may be over 1,000 ohms? This is the extreme example in favor of ESR. What can matter more is dielectric absorption. But only when you generate a signal voltage across the capacitor. By definition this doesn't happen with coupling capacitors.
I've serviced CD players since they were introduced. Trained at Philips and other manufacturers.
Honestly, Electrolytic capacitors are non-critical despite what people say. Each cap varies widely in the same batch for example. ESR may matter for power supply filters, but if this is critical they would use high frequency types. Not common when this was built.
My strong advice? Buy good quality capacitors from authorized distributors. The larger case will be a better part, don't put one in too large. Higher voltage ratings are generally better and help to have the same size case. These capacitors will not change the sound. Others will, but not many here talk about what actually does matter.
I've been a paid technician for almost 50 years, servicing longer. Studied parts since I began, and I have the equipment to do it. Serviced real high end audio and recording studio equipment. I'm just trying to save you effort and money while suggesting the best outcome for you. What you do is totally up to you. For many, parts are a religion based entirely on beliefs and anecdotal reports without real testing. "White papers" are just advertising copy with some technical stuff thrown in that often isn't applicable or truthful (it's an ad, remember?).
I'll say it one more time. There is no best make / model of capacitor. Each part has characteristics that make it better for one circuit location over others, and production tolerance makes the difference between good brands and models insignificant. ESR has no bearing in coupling applications, except in series with a load (higher current). What differences matter even at one ohm when the circuit impedance may be over 1,000 ohms? This is the extreme example in favor of ESR. What can matter more is dielectric absorption. But only when you generate a signal voltage across the capacitor. By definition this doesn't happen with coupling capacitors.
I've serviced CD players since they were introduced. Trained at Philips and other manufacturers.
Um, I read it slightly differently. What was important was that the signal voltage across the electrolytic had to be kept to a minimum (typically below 80mV) with no DC bias for low distortion to apply. He tested down to 6.3 volt electros. This meant that electrolytics in the feedback path couldn't be used to define the lower frequency limit.
Keith
I think that's what I said too. I fully agree with Douglas on this, it has also been substantiated by others; Cyril Bateman already found this out decades before Self. The higher the signal voltage across the cap, the higher the distortion.
As a cap's impedance rises with falling frequency, the signal voltage across the cap rises with falling frequency (in the position of feedback path shunt leg). Therefor, the distortion rises with falling frequency.
It's all a matter of engineering and how important you feel it is and how much effort/money you want to spend, certainly you cannot say that they 'couldn't be used'. They are used in many great amps.
Edit: this is all part of Douglas' efforts to design a 'blameless amp', which, as many incorrectly think, is not a perfect amp. It is an amp where all the usual error sources are minimized. Carefully selecting this feedback shunt elcap with regard to capacitance and voltage spec was part of that optimization process. This actually works; selecting the cap based on a label stating 'special audio cap' does not 😎
Jan
As a cap's impedance rises with falling frequency, the signal voltage across the cap rises with falling frequency (in the position of feedback path shunt leg). Therefor, the distortion rises with falling frequency.
It's all a matter of engineering and how important you feel it is and how much effort/money you want to spend, certainly you cannot say that they 'couldn't be used'. They are used in many great amps.
Edit: this is all part of Douglas' efforts to design a 'blameless amp', which, as many incorrectly think, is not a perfect amp. It is an amp where all the usual error sources are minimized. Carefully selecting this feedback shunt elcap with regard to capacitance and voltage spec was part of that optimization process. This actually works; selecting the cap based on a label stating 'special audio cap' does not 😎
Jan
Last edited:
I get several years ago various information from Mr. Jens Both - developer at Philips Components - later BC.
in post #2234/pg112, and in the posts on pg148 here you will find several basic advices.
In case of the Philips vintage CD player CD303 there are to find a wide range of blue Philips electrolytic miniature capacitors in an axial outline - go to the images under
https://www.audiovintage.fr/leforum/viewtopic.php?t=109&start=110
here I recommend to replace with foil versions all parts by a value of 3µ3 and below like 2µ2, 1µ5, 1µ, µ68, µ47 and µ22 (e. g. WIMA MKS). Very important around the APC circuit for the laser diode.
Additional (at least in my experience), it's best to avoid replacing low voltage capacitors (3V-6V-10V-16V-25V-35V) with capacitors of the same voltage. It's better to replace them with either 50V or 63V capacitors (halve the capacitance values if necessary - there's no disadvantage in replacing a 2200uF/35V or a 4700uF/16V capacitor by a 1000uF/63V capacitor). This increases reliability and lifetime.
At least 105 degree versions should be used on the PCB with the outdoor heat sink.
Panasonic and Nichicon are good quality and easy ton find,, my favorite brand are FTcap and Exxelia (SIC SAFCO)
https://exxelia.com/en/list/capacitors/aluminum-electrolytic-capacitors/axial
https://exxelia.com/storage/exxelia-assets/datasheets/cubisic-v1.pdf
😎
Jan
Bias is another issue, separate from the caps. Bipolar elcaps are usefully in some situations and basically is two series caps in opposite polarity. Again nothing to do with voltage dependent distortion, they ALL have that.
Jan
Hi tiefbassuebertr,
After servicing these things for years, I can say this.
The original capacitors weren't very good quality (big shock). My very strong advice would be to replace them with the same package capacitors. Use the highest voltage that will fit. Simply using a decent brand will allow the machine to last. Don't mess around with values and types of capacitors / capacitance. Anything would be better than the original Philips capacitors!
Fault number two. Gribblets, a Philips answer to how to make the PCB cheaper. These top to bottom connector fail, the reason is the solder didn't wet originally. Why? Cheap production. Remove the original solder, apply liquid solder flux, resolder. Do not use too much solder, you only make a blob that may still not wet properly.
I can't tell you how many of these I see with cheap radial caps installed. Yank those out, install the proper ones..
After servicing these things for years, I can say this.
The original capacitors weren't very good quality (big shock). My very strong advice would be to replace them with the same package capacitors. Use the highest voltage that will fit. Simply using a decent brand will allow the machine to last. Don't mess around with values and types of capacitors / capacitance. Anything would be better than the original Philips capacitors!
Fault number two. Gribblets, a Philips answer to how to make the PCB cheaper. These top to bottom connector fail, the reason is the solder didn't wet originally. Why? Cheap production. Remove the original solder, apply liquid solder flux, resolder. Do not use too much solder, you only make a blob that may still not wet properly.
I can't tell you how many of these I see with cheap radial caps installed. Yank those out, install the proper ones..
I'm a bit older, so I don't know if it still matters these days. The dielectric also determines the distortion.
Hi Franz,
Dielectric does for sure, but in electrolytic capacitors it is very similar between manufacturers. Excluding the "problem" the world saw in the 1980's with a copied formula that was incorrect. That's why various electrolytic capacitors will perform very similarly. The same can be said for Polypropylene or any other base type.
There is a difference between electrolytic capacitors (designed for high frequency switching applications) and your regular electrolytic capacitor. But like everything, you have to look at the data sheet and understand all the characteristics of a part. Not just one that sounds good to you. You have to know your parts.
It still matters.
Dielectric does for sure, but in electrolytic capacitors it is very similar between manufacturers. Excluding the "problem" the world saw in the 1980's with a copied formula that was incorrect. That's why various electrolytic capacitors will perform very similarly. The same can be said for Polypropylene or any other base type.
There is a difference between electrolytic capacitors (designed for high frequency switching applications) and your regular electrolytic capacitor. But like everything, you have to look at the data sheet and understand all the characteristics of a part. Not just one that sounds good to you. You have to know your parts.
It still matters.