Did the B+ change? Did the ripple increase? Did you put a scope to it to see if there was a problem, ..hum.. Did it really cost $0.75? What did a common electrolytic cap ...sound... like in that position? Unfounded hyperbole is what really kills the interest in posting these days.
The historic track record of most 'cap threads' is that they end up getting closed and for the very reasons that this thread has come to our attention.
You have all had your say at this point so please keep it civil and also bear in mind that any sweeping comments such as losing bass response following a component change is more than likely to be followed by asking for objective measurements on what occurred.
B+ was the same, no extra ripple or hum. I have built this same circuit using an electrolytic cap and it sounds fine. I thought it sounded a little cleaner when I used a Solen film cap as the first cap in the power supply, so I have been using them. Honestly that is debatable and I wouldn't post that this is an audible difference easily heard. Before this happened I honestly didn't believe the power supply would impact the sound in an SE amp beyond hum from too much ripple in the DC.
What happened was: I have been using this amp for several years and then one day on power up the rectifier tube flashed, and blew the fuse. I discovered I had pinched the solen cap causing it to short out, which killed the rectifier tube. I looked in my parts bin and saw some large DC-Link Wima film caps I had bought for another project but they didn't fit. I thought to myself "well this will be an improvement if nothing else in durability", they were withing a few % of the one I was replacing and had aprox the same voltage rating, I installed the cap, powered the amp, checked the B+ voltage and was a happy camper.
Hooked it back up to my system and was listening to some familiar music and noticed there was a big change in the bass response. I thought I had somehow messed up some other wiring in the amp as it was not subtle. Remember this was my daily use amp I have listened to for years, I know what it should sound like. Took it back to the bench, all the bias and other voltages looked good and I was scratching my head. It didn't make any sense to me that changing this power supply cap would do that. So went back and listened again, then swapped in another amp I always felt sounded similar and indeed there was a pretty drastic reduction in the bass response on the amp I swapped the Wima cap in.
So knowing the only thing I changed was that one cap, I went online and ordered another replacement Solen cap. When it came in, I once again listened to the amp, powered it down, swapped the cap and the difference was immediately obvious. I thought "Yes! My amp is back". I was baffled by this and even bought an impedance breakout board for my analog discover 2 to test the impedance vs frequency of the two caps. They tested identical. The measured uf was 30.5uf for the Wima and 32uf for the solen, not enough to account for this change. The construction materials and specs all matched as well.
I have no bone to pick and honestly never thought of Solen caps as being some top tier audiophile part. I went into this thinking the Wima cap would actually be higher quality and still am not sure what caused this. If you don't want to believe me, or think I'm just making all this "unfounded hyperbole" up.. I honestly have better things to do than make up something like this. I just wanted to share this experience as it was something I never expected and still doesn't make sense. I have learned in life that the more we know, sometimes just highlights how much we really don't know.
I'm done. PM me if you have any further questions.
I often use Wima MKP10 and Sprague 716 (orange drop) as coupling capacitors on my tube amplifiers because they are popular for this application, easily available, and comparatively small and cheap. They are worth a try.
I once used a DC Link capacitors (not a Wima one) as power supply first filter capacitor because more than 500V was required and the chassis had enough space, but I try to avoid physically large capacitors, expecially inside crowded chassis, because on my experience they may introduce unwanted couplings. Placement becomes important and things starts to randomly change behaviour, often for the worst, just by moving parts around or because a certain capacitor is square and another one is round. They may measure exactly the same on the bench outside the amplifier, but they obviously interact in a different way with other electrical or magnetic fields inside the chassis. This is well known on radio frequency applications. Unfortunately, any digital circuit or switched mode power supply near our analog tube amplifiers is a radio frequency source, a powerful one sometimes.
I once used a DC Link capacitors (not a Wima one) as power supply first filter capacitor because more than 500V was required and the chassis had enough space, but I try to avoid physically large capacitors, expecially inside crowded chassis, because on my experience they may introduce unwanted couplings. Placement becomes important and things starts to randomly change behaviour, often for the worst, just by moving parts around or because a certain capacitor is square and another one is round. They may measure exactly the same on the bench outside the amplifier, but they obviously interact in a different way with other electrical or magnetic fields inside the chassis. This is well known on radio frequency applications. Unfortunately, any digital circuit or switched mode power supply near our analog tube amplifiers is a radio frequency source, a powerful one sometimes.
It's no more bold than the praise poured over the Wima by another poster, who wasn't asked to back up his statement, but I digress.
Take a look in Merlin Blencowe's excellent book "Designing High-Fidelity Tube Preamps" (there is no addition given, but mine has 2016 printed in the beginning). On page 79 in this book is a diagram with comparative tests of nine different film caps with frequency on the x-axis and THD+N on the y-axis. The Wima comes in dead last. To quote the text, it says "The worst performer in fig 2.10 was a Wima red box that apparently suffers from two sources of distortion: the first is the same as that exhibited by all capacitors while the second is proportional to the power flow in the device, producing the pronounced hump in the graph".
Interesting. Specifically, WHICH Wima film cap is he referring to? They have many different models of film caps. And what was the capacitance and voltage value? Wima's best are the film/foil FKP2; however, these are only available up to ~22nF. The next best is their FKP10. AFAIK, "red box" can refer to any of Wima's caps from polyester to metalized polypro to PET to poylpropylene---and these can have VERY different distortion characteristics.
I believe it was 50V, no idea of capacitance or series. The test is presented under a plastic film dielectric capacitor heading, so presumably that is what it is. Further, the Wima was considered to fit in the group before the test by the author, so it stands to reason it's poor performance was a surprise, but that is speculation on my part.
I encourage you to read the book and maybe draw your own conclusions. If you have the equipment, maybe even conduct your own experiments, or you can do what I do, don't worry about it. I use cheap Panasonic polyester caps in my circuit boards, carbon film resistors in the signal path and MOSFETs, TL431s and BJTs sprinkled throughout, and my amps sound fantastic! That said, that's not the recipe for many, so YMMV. This is the internet after all.
I encourage you to read the book and maybe draw your own conclusions. If you have the equipment, maybe even conduct your own experiments, or you can do what I do, don't worry about it. I use cheap Panasonic polyester caps in my circuit boards, carbon film resistors in the signal path and MOSFETs, TL431s and BJTs sprinkled throughout, and my amps sound fantastic! That said, that's not the recipe for many, so YMMV. This is the internet after all.
Last edited: