Cyrus two - non polar cap

Hi,
Last night i horridly found out that c43 (470uF 6v DC feedback filter cap ?) has gone bad. (Swollen and cracked on both channels). I need to replace them both of course.
However, i do not have such easy access to non polar caps where i live.
I am wondering if it would be too horrible to replace it with a polarized capacitor (the neg pole being connected to the 1k resistor).
Can it harm anything or Deteriorate the sound quality?
Will it effect the life span of the cap ?

P.S.
There are quite a few sources for the service manual on the net, some of them specify that this cap is polar, some don't (meaning its non polar).

Thanks -
Liquias
 

Attachments

  • cyrus1pwr.gif
    cyrus1pwr.gif
    22.8 KB · Views: 606
These are the infamous red roederstein caps - good sound but not for eternity.
I am pretty sure that you can replace them with polar elcaps, like it is done in so many other amps. Panasonic FC with 470yF/6.3V are available at rs-components, Vishays at farnell Israel.
 
I agree

It does, seem that my amp is suffering from old age symptoms. I guess all Cyrus two models will suffer the same symptoms soon, none of them are very new these days...

I will look for the panasonic or vishay caps ASAP, I work allot with Farnell Israel.

I wasn't sure why a non-polar cap was chosen in the first place, as far as i know non-polar caps usually play an important role when polarity changes allot and regular polar caps would boil and burst for half of the time... I must say, i can't see how this is the case in this example.
Is it a DC feedback filter cap ? Why were the originals non-polar ?

Dear lohk and pokoyo:
Thank you guys (lohk, I seem to be thanking you allot these days :) ).
I will replace the other two caps mentioned above.
I do prefer not to use black-gates, I don't mind the price but I read they are out of business and any stocks anyone carries of BG is an old stock. I'm sure I will be able to find some other good quality caps, Please let me know if you think this is a mistake...
 
I experienced that not only the value of a cap is important but somentimes also the voltage. In several big studio consoles it makes a significant difference to use the right value: Low voltage elcaps in non-biased circuits sound much better.
Thats why Henry Azima did "voice" the Cyrus amps deliberatly with these low voltage bipolar elcaps. So maybe the low voltage is more important for the sound than if it is polar or not. I should start a thread about this sometimes.
The Cyrus amp also uses 2y2F bipolar elcaps as input capacities. A good quality foil cap should be better here. But maybe this was also a deliberate decision.
 
Ex-Moderator
Joined 2002
Hi, if they are cracked I suppose they were chosen too low in voltage. Normally only the caps that are directly at the power supply lines crack or fail. In other words: the big filter caps. It won't hurt to use non polar 16V electrolytic caps and they won't crack because of overvoltage.

BG does not deteriorate as standard caps do. If you can get hold of them you probably have the best caps you can use at that spot. However, I think they will be way too large physically. Another very good technical and musical choice is Nichicon MUSE ES non polar caps. They won't break the bank with $ 0.72 each !

http://www.hndme.com/productcart/pc/viewPrd.asp?idproduct=747&idcategory=40

I also would like to suggest to replace all electrolytic caps as they are quite old now. There is nothing worse than a good repair followed by the same error of another cap. The total costs are low anyway and it will eventually cost less time. And if you don't replace all caps regret will follow if a bad caps takes some semiconductors with it to the electronic nirvana. I would replace C31 and C57 for Wima MKS2 5 mm pitch polyester film caps. As always BG N series 2.2 or 4.7 uF is a good but expensive choice here. You can replace all polar caps by Pana FC or the likes. You really can't go wrong with Pana FC/FM. You can use somewhat larger values for C53 and C55 if they fit physically.

Even if Azima did "voice" the amps please bear in mind that this was in 1988. Caps have changed and you won't find the same ones. Even if you do they will also be too old to use.
 
Last edited:
hi, according to the Cyrus 2 service manual that I checked, the 470uF 6v DC feedback filter cap is not install. it is missing in the circuit diagram.
this 470uF 6v DC feedback filter cap is presence in the Cyrus 1 only.


Hi,
Last night i horridly found out that c43 (470uF 6v DC feedback filter cap ?) has gone bad. (Swollen and cracked on both channels). I need to replace them both of course.
However, i do not have such easy access to non polar caps where i live.
I am wondering if it would be too horrible to replace it with a polarized capacitor (the neg pole being connected to the 1k resistor).
Can it harm anything or Deteriorate the sound quality?
Will it effect the life span of the cap ?

P.S.
There are quite a few sources for the service manual on the net, some of them specify that this cap is polar, some don't (meaning its non polar).

Thanks -
Liquias
 
C43: 470uF/6V3:
100uF is enough for sufficient low cutoff frequency; I use 100uF/63V about
Evox-Rifa | Passives | Capacitors | Aluminium | Axial 105°C |PEG124MD310VQ
and 1uF/63V for bypass about
WIMA | Passives | Capacitors | Plastic Film | Metallised 100°C |MKS4/1.0/250/10
If you have small loudspeakers (bookshelf) use only 1x, 2x or 3x10 uF unipolar foil cap, e.g. EPCOS about
Epcos | Passives | Capacitors | Plastic Film | Metallised 125°C |B32523Q0106J
1uF for bypass now no longer necessary.

C31: 2,2uF: replace this unipolar elcap through foil type
Vishay | Passives | Capacitors | Plastic Film | Axial 110°C |MKP1839522634HQ

Independend of the condition replace additional all elcaps arround the voltage regulators 7815 and 7915 through new versions, at best also through 63V types

These are the infamous red roederstein caps - good sound but not for eternity.
This caps are very good known for its can decomposition, even by audio devices from Krell. If you are find this caps anywhere, you must still replace it through high quality versions and metal can enclosure, also in cases, where the outside condition seems to be good like the attached photos

Visit this website from Mr. Jens Both, if you want more information about the reasons of "bad caps"
http://www.elcap.de/index2.html
 

Attachments

  • roe22-40.jpg
    roe22-40.jpg
    19.1 KB · Views: 353
  • roe 2009-0109-Roederstein 교체 002-1.jpg
    roe 2009-0109-Roederstein 교체 002-1.jpg
    174.9 KB · Views: 347
Last edited:
@jamal,
I have several versions of cyrus two amps, in all of them this cap exists. I know I pulled the defective ones out of my amp, so I did have them originally. I would like to see (if its possible), the additional version of the schematic for the amp.

@tiefbassuebertr,
I changed the caps to the non-polar nichicon muse 470uF 16V.
I can't seem to find 470uF 6V non polar caps, from a good brand.
I tested polar elna cerafine 470uF 6.3V sounded very good but they get hot for some reason I can't understand right now. I measure no DC across them, and I don't know why they heat up.

I'll try and post more results and questions tomorrow.
Thanks guys!
 
hi Liqulias, the cyrus 2 schematic I refering to, is issue 2006. the whole service manual is copyright 1999. here are no 470uf 6v non polar capacitor on the feedback path. I intend to put 2 1000uf connected back to back to get a value of 500uf non -polar cap on my Cyrus 1.

@jamal,
I have several versions of cyrus two amps, in all of them this cap exists. I know I pulled the defective ones out of my amp, so I did have them originally. I would like to see (if its possible), the additional version of the schematic for the amp.

@tiefbassuebertr,
I changed the caps to the non-polar nichicon muse 470uF 16V.
I can't seem to find 470uF 6V non polar caps, from a good brand.
I tested polar elna cerafine 470uF 6.3V sounded very good but they get hot for some reason I can't understand right now. I measure no DC across them, and I don't know why they heat up.

I'll try and post more results and questions tomorrow.
Thanks guys!
 
Last edited:
I changed the caps to the non-polar nichicon muse 470uF 16V.
I can't seem to find 470uF 6V non polar caps, from a good brand.
I tested polar elna cerafine 470uF 6.3V sounded very good but they get hot for some reason I can't understand right now. I measure no DC across them, and I don't know why they heat up.
avoid non polar caps with low voltage. Use 63V type. in reverse direction the voltage is still 1V by electrolytics independend of the nominal reading voltage on the device. Check other sources for hot spots on your PCB near the capacitor and check high frequency oscillation with your oscilloscope accross the cap.
Avoid caps "special made for audio" and use only high quality parts for professional applications.
 
Last edited:
in reverse direction the voltage is still 1V by electrolytics independend of the nominal reading voltage on the device.

This is correct for polar electrolytics but non-polar electrolytics have the specified voltage in both directions (hence non-polar). I presume you wanted to say this but it did not come out clearly in your post.

Use 63V type.
Why? Other than higher voltage electrolytics generally having better specs than lower voltage types there is still the issue of size (and often increased inductance). Any supporting points for exactly 63 Volts?
 
This is correct for polar electrolytics but non-polar electrolytics have the specified voltage in both directions (hence non-polar). I presume you wanted to say this but it did not come out clearly in your post.

Why? Other than higher voltage electrolytics generally having better specs than lower voltage types there is still the issue of size (and often increased inductance). Any supporting points for exactly 63 Volts?

This is right. However - non polar caps I cannot recommend, because it needs to large space in case of high quality versions (e. g. Intertechnik "Elko glatt") or its ESR value is too bad. ELNA BP CERAFINE series e. g. is a bad choice. Actually internal two polar electrolytic caps are connected anti-serial by N.P. electrolytics (outside connectors neg. poles - as I know).
The reason for 63V types is the fact, that most brands only by this voltages reaches their lowest value for ESR both at dc and 10/100KHz. This I have heard from a developer. You can see also this in the data sheets by the ESR values by the various VDC values.
By few certainly models lowest ESR is to find at 50VDC or 100VDC, but mostly it is by 63VDC except various brandnew types for PC Motherboards e. g. from Sanyo; go for PDF download to
http://www.endrich.com/de/site.php/110
I don't know why it is just 63VDC Ask Mr. Jens Both - http://www.elcap.de/4437/index.html
read also this threads:
http://www.diyaudio.com/forums/soli...s-electrolytic-capacitors-audio-circuits.html
http://www.diyaudio.com/forums/parts/151392-best-electrolytic-capacitors.html
http://www.diyaudio.com/forums/soli...ter-coaching-certainly-electrolytic-caps.html
 
Last edited:
Well both polar and non polar electrolytics are internally two caps that are in series ("connected via the electrolytic"). The only difference is that in polar caps one aluminium foil is formatted so that the oxide layer grows thicker and withstands the rated voltage. The other is left unformatted and has only the natural thin oxide layer that forms on aluminium under atmospheric conditions and can withstand some 1.4 Volt or thereabouts (reverse voltage before breakdown). For non-polar caps the second foil is also formatted so that is has also a oxide layer thick thick enough for the rated voltage. This might actually explain why they are a bit bigger (or s I assume).

Non-polar caps still show much lower distortions as coupling caps, especially at lower frequencies, so that you will get away with much less capacitance compared to polar caps. So if you have to use a DC blocking cap in such a position the non-polar one may still be the better choice. The non-etched "glatte" caps are again much bigger so I doubt they would fit everywhere.

I do agree though that higher voltage caps have better ESR but I wonder if less voltage would still be "good enough" under most circumstances.

P.S. I just happen to have some old non-polar red roedersteins caps and are wondering if and how you can reformat non-polar caps (reformatting one foil might weaken the other, and vic versa). I will try your contact at Elko-Consulting to see if they have any info.
 
Last edited:
Well both polar and non polar electrolytics are internally two caps that are in series ("connected via the electrolytic"). The only difference is that in polar caps one aluminium foil is formatted so that the oxide layer grows thicker and withstands the rated voltage. The other is left unformatted and has only the natural thin oxide layer that forms on aluminium under atmospheric conditions and can withstand some 1.4 Volt or thereabouts (reverse voltage before breakdown). For non-polar caps the second foil is also formatted so that is has also a oxide layer thick thick enough for the rated voltage. This might actually explain why they are a bit bigger (or s I assume).

Non-polar caps still show much lower distortions as coupling caps, especially at lower frequencies, so that you will get away with much less capacitance compared to polar caps. So if you have to use a DC blocking cap in such a position the non-polar one may still be the better choice. The non-etched "glatte" caps are again much bigger so I doubt they would fit everywhere.

I do agree though that higher voltage caps have better ESR but I wonder if less voltage would still be "good enough" under most circumstances.

P.S. I just happen to have some old non-polar red roedersteins caps and are wondering if and how you can reformat non-polar caps (reformatting one foil might weaken the other, and vic versa). I will try your contact at Elko-Consulting to see if they have any info.

Interesting facts. I have also often heard, that non-polar caps should still have much lower distortions than polarized caps at lower frequencies.
Denon uses even polar caps in their baxandall-equalizer section of the "PMA-1560" (down to 100nF) - go to schemas by post #4 about
http://www.diyaudio.com/forums/soli...-optical-class-pma1560-how-works-circuit.html

Some other developers claim, that tantal caps as coupling caps in the signal pad are the best choice (in case of non symmetric power supplies). Naim Audio is an example. For me the question is, whether this views are really true. Never I have observed a confirmation for that by various listening tests. My experience is, that foil caps in signal pad is always better choice than electrolytics, even in such cases, where the occured unwanted high pass function (and the associated - for perfect stereo panorama unwanted phase shift/group delay time) arises to higher frequencies. An alternative is the use of high quality polarized caps (in case of non symmetric power supplies).
Completly dc coupled stages actually should be the royal way, but this is difficult to realize and any few developers claim, resistors (especially that one for NFB and gain determine) without voltage biased works sometimes like diodes with additional unwanted high-order THD.
So for me the royal way is unsymmetric power supply for audio circuits and either foil or polarized electrolytic caps arround 63V in the in this case naturally no dc coupled signal pad.
 
Last edited: