I tried swapping a pair of amp/preamp line output coupling capacitors (DC blocking). The stock units were run of the mill Panasonic electrolytic. I put in Mundorf Mlytic AG. I know this particular Mundorf (and most large electrolytics) are for power supply use, but the application calls for a large 6800uF so that the system can be used as a small amp or preamp. Since I'm using the unit mostly as a preamp, I guess I could go about 1/10 the size and try to use a film cap.. but anyways, that is besides the point.
The question here is, do electrolytics (specifically this Mundorf) need burn in? I ask because I'm hearing the opposite of what most people hear on caps that are not adequately burned in. The usual complaints with caps without proper burn in are lack of bass, shrill highs, etc. The Mundorfs have increased the bass output and possibly added some texture to the mids (the good), but the highs have disappeared. I'm wondering if the highs will come back over time? I wouldn't say the top is rolled off, but it is just shelved. Any details that are carried in the upper frequencies is obscured and recessed. The life of the sound has been sucked out. Hopefully this doesn't qualify as "audiophile" tuned warm sound.
I have considered putting a small Russian teflon cap in parallel to the Mundorf, but the values I have are too small to probably make a difference (820pF). I also don't think this bypass should be required to get nominal levels of treble in the first place.
I've got about 5hrs on the caps now, and they sound exactly the same as time 0. So are the Mundorf's going to open up the highs with time, or should I just move on to another cap? I will try the bypass before I throw in the towel on the Mundorf. The point of reference is the stock Panasonic caps, which are fairly neutral, with good top end extension and detail.
If new caps in a PSU, will the temporary inefficiency manifest as possibly less responsiveness to peak demands by music at higher listening volumes? In other ways?
The best thing you can do for electrolytic is to cycle them on-off as you would normally in listening sessions. What helps caps is to form the electrolyte (the "burn in") and this is best accomplished by cycling them as you would by normal turn off and turn on of the component as in regular use.
@erik_squires , the preamp is also a 1W power amp, meant to drive 4ohm loads, thus the large cap value. It is the LTA MZ2. Since it was primarily designed as a smallish power amp or headphone amp, in preamp mode, it still likes to be loaded somewhat (between 50-100ohms). Mine is loaded at 100ohms. Therefore, I can easily get away with 1/10 of the stock cap value without an issue (a bulky film if possible). This is something I can look into if the electrolytic doesn't perk up.. The stock cap is not Panasonic FM. It is bluish-purple with a white negative stripe. It is a Matsushita ("M" logo), but I just assumed this was Panasonic.. maybe its not.
@atmasphere , thanks for the tip. How many hours do you think it takes to form an electrolytic? Patience seems to be key..
From Stevecham's comments, it looks like power cycling does the best job of forming the dielectric. I'll put some "don't care" tubes in there and go through the power cycling process, thanks!
@manthik If the caps needing to form up is what is happening, leaving the caps charged is the best way to do it. Cycling it takes longer- weeks or months instead of days.
If new caps in a PSU, will the temporary inefficiency manifest as
possibly less responsiveness to peak demands by music at higher
listening volumes? In other ways?
Yes, and in other ways. We won't even let our stuff out of the shop until it has at least a day on it, because it sounds so 'dead' prior to that. You can measure the form up too- as the caps get more efficient, the voltage on them goes up and they also run at a lower temperature.
@roberjerman, I am using the stock value. There is nothing to second guess here. It is a simple high pass filter network with a primary function to block DC. The value could be changed according to the implementation, if desired. The designer's intention was for the unit to be used as a preamp and amp. Since I only use it as a preamp, I can adjust the value accordingly.
Thanks Atmasphere. I haven't let the unit run in yet, but I'll do that by this weekend and report back.
OK, per suggestions here, I let the amp run continuous for about 48hrs. If you add in my listening hours, the total time would be somewhere about 55hrs. There is very little change in the sound. The bass is still there, mids are nice, but the top end is still shelved (maybe a tad less than before the 48hrs of run time). It sounds akin to listening to the rolled top end of Mullard tube. I'm sure a lot of people would like this presentation. I'm just missing top end details, presence, and bite (when the music calls for it). All of this top end characteristic was present with the stock cap. I'll try to use a smaller film cap in parallel and see what happens. I like the bass and the mids enough with the Mundorf to continue experimenting to see if I can find the right overall compromise.
Leaving a component on continuously will NOT properly form the caps!!! They need to discharge and recharge over multiple cycles to do this. And it is not best to simply turn the component on and off in quick succession. The caps retain charge for several hours once the component is turned off and need to discharge that voltage over the same time frame, which is why cycling on-off, on a daily basis, with an overnight to discharge is by far the best way to form caps.
I have always found the first listen after any change to be a solid indication of what you will end up with. Yeah, burn-in helps but I have NEVER found an initial impression to change from fair to awesome.
Sorry for the slow reply guys. I see there are differing viewpoints as to how to properly form an electrolytic capacitor. I have done some slow rate power cycling, but that didn't really help either. However, I didn't do much, and I'm not sure how much would be required. I basically have hours and cycles on the caps now, and there wasn't any appreciable change in the top registers. Nothing magically changed. I might have to concur with @noromance on this. I have heard profound differences with certain tube burn-in in certain gear though, so I'm fully open to the fact that sound can and does change once properly run in.
Anyways, I added a 3.3uF poly cap (Panasonic WF series) in parallel with the big Mundorf electro just to see what would happen. All the highs came right back. Night and day difference. Almost seemed like the bass is less pronounced too, but I wasn't able to tell yet. After about an hour of listening, my DIY power source for my front end components crapped out. So, it will be a couple weeks before I am back online and can add more impressions. I think this is going to be trial and error based adjustment.
After sorting through all the choices, I finally settled on a huge pair of polypropylene 1000uF caps made for commercial power supply applications (solar power). I found them cheap on ebay. I got them custom mounted. All the highs and the details are there. Very well balanced cap.
The shunt film cap on the large EL never worked over the long term. Over time, the combined cap would sound more like the EL cap (I think the film caps I was using were still burning in). Also, the film cap values were too small relative to the EL cap to make a difference in the long run. As one might expect, no EL cap can stand up to a good film, and that goes for all of the best. For EL caps, I tried Mundorf, Elna Silmic II, Nichicon, etc.
After sorting through all the choices, I finally settled on a huge pair
of polypropylene 1000uF caps made for commercial power supply
applications (solar power). I found them cheap on ebay. I got them
custom mounted. All the highs and the details are there. Very well
balanced cap.
The shunt film cap on the large EL never worked
over the long term. Over time, the combined cap would sound more like
the EL cap (I think the film caps I was using were still burning in).
Also, the film cap values were too small relative to the EL cap to make a
difference in the long run. As one might expect, no EL cap can stand
up to a good film, and that goes for all of the best. For EL caps, I
tried Mundorf, Elna Silmic II, Nichicon, etc.
The value of the part can make a difference. There is more nuance to power supply design than initially meets the eye. If the cap values are too large, the impedance curve of the output of the supply might not be a good bypass at high frequencies even though the overall output impedance might be lower. If too small, bass might not be well bypassed. This can change depending on the performance of the caps used- better caps can often be larger values. But you have to watch your timing constants (as a general rule of thumb, the TC of the power supply should be at least an octave lower than the TC of the audio circuit its supplying, and not much more!) and if you have regulation, many regulators won't be happy with larger values (and might have shorter lifespans too). This is the tip of the iceberg! I would not write off conventional electrolytics though; if used correctly they can perform as well as film caps. If. Used. Correctly. YMMV but IME we've had terrible luck with using more exotic electrolytic capacitors; they don't seem to work as well as more middle-of-the-road parts. Quite often they wind up sounding dark.
Hi @atmasphere , the cap I was discussing was actually being used for output coupling in an amp/preamp. Stock value is very large, because the circuit is meant to drive a speaker load. I noticed the Mundorf EL cap in this location was inherently dark sounding (no highs). Anyways, thanks for the comments. Im currently happy with the large polypropylene film caps that are in use (havent touched them). At 1000uF and high voltage rating, they are as big as the chassis for the amp haha
Once the electrolytic has formed up, it should work about as well as the film caps in a coupling cap situation, so long as its biased correctly. To this end, whatever AC voltage is being sent through it as an audio signal should never exceed the DC voltage on the cap. I don't know how a coupling cap in a traditional solid state output could do that- The DC on the cap would be 1/2 of the power supply voltage and the output swing of the amplifier could be nearly rail to rail. So in such a circumstance the cap would make considerable distortion. I think this is one of the ways electrolytics have gotten a bad rap in audio as coupling caps.
That is true. My preamp puts out about 6v DC, and the AC signal component would have to exceed 2Vrms to cause the polarity to flip on the cap. That kind of level would exceed my amplifier sensitivity by 2x. Needless to say, I dont listen very loud, so in my application, I could safely say the polarity of the EL cap was never dynamically reverse biasing. Thus, it had no bearing on the sound I was hearing. All of the EL caps I tried (after lots of hours to form) were all lacking detail and lacking HF content. This was a common theme, although some fared better than others.
You must have a verified phone number and physical address in order to post in the Audiogon Forums. Please return to Audiogon.com and complete this step. If you have any questions please contact Support.
