Capacitors I'm not familiar with...

[Boomzilla]

With shelter in place being the norm these days, I plan to do a bit of soldering. My old cache of Sprague orange drop caps is mostly gone now, and I'll need to buy some new capacitors. But the brands I find for sale are none that I've ever used before. Choices seem to include:

Hilitichi
Totot
Wimas
Wingoneer

Are there any of these brands you'd recommend (or any to stay away from)?

Thanks - Boomzilla

[DYohn]

What type of cap? What are they for?

[Boomzilla]

Looking for polyethylene dielectric caps of about 200-volt rating. I'd be using four of them to make a balanced outboard high-pass crossover of first-order slope. I'll need four capacitors, but would like to buy about 20 so I can hand pick the values closest to my target of 0.072343 μF.

[DYohn]

For crossover work I like Mundorf, Cross-Cap, Solen, Jantzen, Clarity, and I have had great sounding performance with some generics out of Germany. Use film caps if you can, and failing that metalized poly. You don't need 200 volt rating unless this is for a very high power application (like 5000 watts amplifier power. You are not doing that.) Buy higher tolerance if you want higher quality. 1% caps cost 10X 10% caps, in general, but they will perform closer to their rating - and far more importantly closer to each other. I don't know how you got that 0.072343 microfarad number because that would create an ultrasonic high-pass filter up around 250KHz. Definitely not in the audible range.

So, perhaps a better question, what are you trying to build? If this is for a loudspeaker, what are you trying to do?

If you are building a radio or something than I am not your man.

[bolle]

Googling the Sprague Orange Drop could something like the Silver Mica be a good replacement? I used those so far as bypass caps with good results.

[ttocs]

Apr 1, 2020 9:00:58 GMT -5 DYohn said:

 I don't know how you got that 0.072343 microfarad number because that would create an ultrasonic high-pass filter up around 250KHz. Definitely not in the audible range.

Maybe he's connecting them in series?

edit: scratch that, parallel makes more sense. but what do i know? it's a puzzle.
also, it's April 1st. ya never know . . .

[donh50]

Not sure they make the ones you need, but commercial caps I have used include Murata, Panasonic, TDK -- check DigiKey, Newark, Allied, and those kind of places. DYohn provided a good list of "audiophile" caps. I would choose poly film in some variant, then maybe non-polar electrolytics for large values that will be less expensive. I would avoid ceramic, tantalum, regular electrolytics, and mica for various reasons (technical, can't really speak to their "sound").

[KeithL]

Mica capacitors are notorious for producing audible amounts of distortion.

Electrolytics can be similarly poor in sound - but it depends on exactly how they're used.

Ditto for tantalums and ceramics.

Many people also choose to bypass large electrolytics (for example, a 100 uF electrolytic paralelled with a 10 uF film cap).

Good quality film caps are often worth it... but avoid uber-expensive "audiophile capacitors"... they're never worth the ridiculous prices they sell for.

Dayton audio makes quite nice metallized polypropylene and film caps - and their prices are reasonable (Parts Express).

Apr 1, 2020 16:42:33 GMT -5 donh50 said:

Not sure they make the ones you need, but commercial caps I have used include Murata, Panasonic, TDK -- check DigiKey, Newark, Allied, and those kind of places. DYohn provided a good list of "audiophile" caps. I would choose poly film in some variant, then maybe non-polar electrolytics for large values that will be less expensive. I would avoid ceramic, tantalum, regular electrolytics, and mica for various reasons (technical, can't really speak to their "sound").

[KeithL]

Bypass caps in applications like power supplies are usually small values - and are only intended to have very low impedance at VERY high frequencies.

(At lower frequencies the main capacitor does all the work... but most larger capacitors have limited performance at very high frequencies.)

Apr 1, 2020 15:55:58 GMT -5 bolle said:

Googling the Sprague Orange Drop could something like the Silver Mica be a good replacement? I used those so far as bypass caps with good results.

[audiobill]

Just be sure the caps aren”t the blow up brands some component brands favor. Cheap, but at what cost?

[donh50]

I happen to be working on a PDN (power distribution network) for a product (way, way beyond the audio band) and am correcting a problem caused by somebody who decided to just throw a bunch of capacitors and inductors of various sizes down based on "it's what we've always done". Aaarrrggghhhh...

As Keith implies, capacitors and inductors are not ideal, and will resonate and then quit doing their job past a certain frequency. That is, at high frequency a capacitor looks like an inductor, and inductors start looking like capacitors. Put the wrong combination together and you can get funny resonances that cause problems. And even if not, more is not always better... I had to hold my tongue when a customer (long ago and far away, but in this galaxy) proudly showed off the massive set of capacitors bypassing a woofer cap in a crossover network (that shunted HF energy to ground before it hit the woofer). It probably did no harm, but the guy spent like $100 on a bunch of caps that would provide fairly effective bypassing to about a GHz (if he had better soldering skills and didn't leave such long leads on them). Since woofers rarely see more than few kHz even in a two-system (his was three or four, I forget) RF bypassing was expensive and unwarranted. IMO -- he liked the sound, natch.

A little knowledge...

[bolle]

Apr 1, 2020 16:55:01 GMT -5 KeithL said:

Bypass caps in applications like power supplies are usually small values - and are only intended to have very low impedance at VERY high frequencies.

(At lower frequencies the main capacitor does all the work... but most larger capacitors have limited performance at very high frequencies.)

Apr 1, 2020 15:55:58 GMT -5 bolle said:

Googling the Sprague Orange Drop could something like the Silver Mica be a good replacement? I used those so far as bypass caps with good results.

I was talking about use as a bypass cap for a highphass filter in a passive speaker crossover. Yes, I also use tinfoil caps and paper-oil and stuff like that, because to me it DOES sound better in a blind, level matched A/B comparison.

I won´t give any advice or talk about use in electronics, because I don´t know anything about that. Only to not buy crap and rather overreserve regarding voltage and temperature than have no headroom.

[Boomzilla]

This is to be a line level filter - not at the speaker level. Power amplifier input resistance is 23,500 ohms (per leg of a balanced connection). Calculation for a 100-Hz. high-pass (single pole) filter is 0.068 microfarads. Should I bother doing this, I'll probably also rig a third-order Butterworth low-pass filter for the subs. That calculation indicates I'd need coils of 28,059mH and 56,094mH with a cap to ground in-between of 0.09 uF. This assumes the same 23,500 ohms (per leg of a balanced connection), but the subwoofers' input resistance is yet to be confirmed.

So the overall parts list for this "line-level" balanced-crossover would be:

4 film caps of 0.068 mF (first order high-pass) and the remainder for the low-pass, 3rd-order:
4 coils of 28,059 mH
4 coils of 56,094 mH
4 caps (NP electrolytic would be fine) of 0.09 uF

This would be totally a totally passive crossover with no active amplification stages. However, parts variation may make this infeasible. There's a limit to what values can be had and to what accuracy. It may end up being cheaper to fix my Dahlquist DQ-LP1 that uses a single-pole passive high pass and a IC-Op-amp based third-order low pass. It also prevents me from having to create my own project box.

[DYohn]

For line-level filters you are likely better using active circuitry due to tolerances of passive components at low line voltages making huge differences in performance. Or just buy a mini-DSP.

[Boomzilla]

Thanks, David - what I suspected... Since I've already got the Dahlquist, I'll just focus on repairing that.

[DYohn]

What Dahlquist? The DQ-LP1? You do realize that is intended for dding a subwoofer to one of their speaker systems, so it isn't the most flexible device. What are you trying to do?

[repeetavx]

For your consideration and entertainment.

Audiophile Caps

[simpleman68]

Apr 1, 2020 9:00:58 GMT -5 DYohn said:

For crossover work I like Mundorf, Cross-Cap, Solen, Jantzen, Clarity, and I have had great sounding performance with some generics out of Germany. Use film caps if you can, and failing that metalized poly. You don't need 200 volt rating unless this is for a very high power application (like 5000 watts amplifier power. You are not doing that.) Buy higher tolerance if you want higher quality. 1% caps cost 10X 10% caps, in general, but they will perform closer to their rating - and far more importantly closer to each other. I don't know how you got that 0.072343 microfarad number because that would create an ultrasonic high-pass filter up around 250KHz. Definitely not in the audible range.

So, perhaps a better question, what are you trying to build? If this is for a loudspeaker, what are you trying to do?

If you are building a radio or something than I am not your man.

Just here for the read. I was proud of myself for being able to comprehend about 10% of the material.

Scott

[Boomzilla]

Apr 2, 2020 9:26:36 GMT -5 DYohn said:

What Dahlquist? The DQ-LP1? You do realize that is intended for dding a subwoofer to one of their speaker systems, so it isn't the most flexible device. What are you trying to do?

Although the original design intent was specifically to mate the DQ-10 speaker to the DQ-1W subwoofer, the crossover is deliberately intended to be flexible enough to work with a wide variety of speakers and subs. The first order slope on the high-pass leg minimizes phase issues at the crossover point and maintains the same phase as the 18 dB / octave slope on the subwoofer(s). The device has the additional advantage of allowing discreet right and left channel or summed bass. Further, even the passive high-pass section can be modified without soldering by a clever set of resistor-holders on the board that can be used to change the high-pass crossover frequency without needing to actually change the capacitor. One can add resistance in parallel or in series with the amplifier's input resistance to tailor the frequency as desired. The low pass section, being active, can be adjusted from the front panel to match. The Dahlquist DQ-LP1 is supposedly considered one of the best-sounding and most flexible crossovers ever designed for home audio (the JL Audio CR-1 perhaps being the sole exception).

At least one of the op amps in my left channel has died (a not unusual failure mode for this item, apparently), and if I'm going to replace it, I might as well use a socket. But it'll still be easier to repair this crossover than to build another from scratch.

[DYohn]

You've obviously done some homework on this thing, so kudos and go for it getting it fixed. A DSP crossover would be far more flexible. But, if all you want to do is integrate a subwoofer into a stereo setup, you really don't need more than a dedicated subwoofer amp that has phase, low-pass and level controls. I had a sub integrated into my 2-channel system for years with no high-pass on the mains. It's all about what you want to do and how you want to do it.