Press J to jump to the feed. Press question mark to learn the rest of the keyboard shortcuts
ArchivedStickied post

Parts sellers around the world.

ArchivedStickied post

Edited 1-25-16

Living room hi fidelity speaker system. 10 inch 200 watt, parts list, wiring diagram, photos and links.

                          cost per speaker                

FaitalPRO 10FE200 10" Professional Woofer $53.00

Tymphany (Vifa) XT25TG30-04 1" Dual Ring Radiator Tweeter $33.00

(Alternate tweeter, Tymphany NE19VTS-04 3/4" Silk Dome. This is an old favorite of mine that I used in 3-way systems. The small diameter takes up less space. The 3/4" silk dome & neodymium-iron-boron magnet combination provides a bit sweeter sound than the ceramic magnet in the XT25TG30-04. Use the same capacitor and coil listed below and use 3 12 ohm resistors in parallel, instead of 4, in series with the woofer, making a 4 ohm high power resistor to match the SPL outputs of the two drivers.

ERSE 1.0mH 18 AWG I Core Inductor Crossover Coil $7.50

Madisound 12 Ohm, 25 Watt Wirewound Resistor (x4) $0.75 each

Film capacitor Solen, 6.8uF / 700V, AGM Series, Silver Metalized $24.00

Solen Inc. 6.8uF

Terminal Plates $5.50

Acoustic Cotton $30.00 Echo Absorber Acoustic Panel (3 pieces of 1" x 2' x 4')

Cabinet costs vary from what kind of wood you use.

I was asked if there was an easy way to improve this speaker. There is always a way to improve a speaker and using a better capacitor is the easiest way. The MultiCap, PPFXS Polypropylene Film & Tin Foil Capacitor, 7.0ufd 100 volt cap is a better cap because it is only 100 volt and it is oval. Tin has almost the same low resistance as silver. The oval shape makes a big improvement and the lower voltage is clearer. Use a .338uH coil, unwrap 44 turns off the 1.0 Erse coil. The good news is the improvement is noticeable but the downside is the cost at $65.00 each for the caps. Once you get into building high fidelity speakers, a small improvement cost a lot of money. This ’little’ 2 way with the oval cap will deliver all the fidelity a $2500.00 amp has to offer.

Older projects posted here have become obsolete due to parts no longer being manufactured.


Audio classes in tech school don’t have enough time to teach all that there is. My school spent 2 days on loudspeakers and could have easily spent 3 more days and still not covered it all. This sub Reddit has all that I learned in school and in the almost 40 years after. In short, I’m spilling my guts.

I now have the experience and knowledge to build speakers that sound like the +10k high end speakers at a cost between $500.00 and $1,000.00 per pair, with $150.00 per pair going to the oak cabinets and covers. Plywood cost less than half of the oak, with covers. All my ’tricks of the trade’ are in here somewhere.

The technology over the past 30 years has improved the woofers, midrange, and tweeter drivers by a huge amount, and the cost has come down at the same time. The same is true with capacitors, a major component of a high fidelity system.

The format of Reddit puts the newest posting first so there is no special order to the postings. Most of them are a paragraph or two for easy reading. Most of them have some extra information not named in the title and are worth reading. Anyone who reads all of them and understands what they read will be able to go toe to toe with the best educated techs and talk speakers like a pro.

The charts and graphs in the side bar contain composite information from many charts and graphs showing the relationships between many topics. The sticky post has links to just about everything you will ever need to build great speakers at an affordable cost.

Please take the time to save the pics onto your computer. I used Imgur for safety reasons, they are trustworthy. Also copy and paste the written material onto a word document page, or to your email and send it to your self for archiving. Burn it all to a disc for hard backup, just because it is a good idea to do that. (EDIT, 10-12-15: Drudge says that he was told by a Supreme Court Justice that news aggregators like Reddit will be soon be prevented from linking to news stories as copyright infringement. So Reddit may be about to be killed. COPY everything, just in case)

Build yourself a great sounding pair of speakers. Don’t start out buying the highest priced drivers and capacitors expecting to get better sound. You won’t, no one does. The cabinet work is harder to do than installing the parts. That will determine what you charge for your speakers if you want to do that kind of thing.

Building great sounding speakers has always been satisfying and rewarding. I’ve heard the words “Build a pair for me” often and it feels good to hear that. The first few pair I sold I answered "I'll sell you the ones we are listening to". They were pleased to get a pair they knew sounded good and I was ready to go on to my next project.

Edit. Your thanks below are greatly appreciated. Some people don't like to see this amount of info made available to every one, especially for free. Thanks again and I'm glad to have made this sub.


Short answer is you used caps with different voltage ratings. This sounds absurd, but it is true.
(Edit. I use a meter to check my cap sizes and keep them to within 1% of the values needed so tolerance isn't an issue.

We were all taught that the effect of moving electrons is felt at the speed of light. This is true on a conductor. A capacitor is not a conductor, but it does pass AC thru an insulator. Apparently, the electric field passing thru an insulator slows down, then the electrons on the other side of the insulator return the signal to full speed. The slower speed is somewhat constant and the higher the voltage rating on the cap, the thicker the insulator, the longer the time it takes for the electronic field to pass thru the caps insulation. A 200V cap is considered fast, 600V caps are medium, and 1200V caps are slow. A small difference every 200 volts is discernible to the most experienced listeners, most people won't notice it.

I’ve heard of this a few times over the years and thought the guys were nuts because I could not duplicate the results from one speaker pair to another. What I didn’t know was when I built a good sounding pair of speakers with its crossover caps picked for those drivers, I was really using the caps to compensate for differences in the drivers resonant cones, dust covers and sometimes the chamber behind the driver. When moving a crossover to another speaker with different drivers, I got different results. I never expected the drivers themselves to be able to have that much influence over the sound. In my early days I used Bozak and KLH drivers. The small diameter voice coils prevented any cap and chamber resonant problems and the manufactures of those driver made sure the cones didn’t have a resonant output. That is not true these days.

I was still trying to figure out the caps when I started this sub Reddit and have done a lot of looking around for info to put in here. While doing that one day I was looking at, and saw what Robert Hovland of Hovland caps had to say and it made sense. So I started to redo my crossovers and kept the voltage ratings the same and got better results. After reading what Lynn Olson had to say about resonant problems from the dust caps, etc. I went after that problem and talked about it in the “In the absence of resonance….”) post. I dusted off an old idea about using 2 woofers in parallel in a two way and built a few pairs using very different drivers in each pair and got them to sound exactly the same, for the first time in almost 50 years of speaker building.

As technology improved over the years, AM to FM, solid state to IC chips, tape recorded music at the studios to hard drives, more and more noise was removed and that led to hearing the remaining noises, sometimes for the first time. This really puts the remaining burden of high fidelity music reproduction on the speaker system. I’m sure the driver manufacturers will play catch up when they realize they have to. Until then, remove the resonant noises from the woofer and midrange drivers yourself.

Back to the cymbals. If you use several caps to build your crossover, keep them to within 200 volts total. When you have 600 volts or more difference you will get a delayed signal from the higher volt cap, in respect to the other caps, and this will cause cancellation and the frequency of cancellation will peak when the time of one alteration is equal to the time of delay.

I’ve searched for info about this loss of cymbals and found more questions than answers. One answer was that the cancellation was due to the placement of the speakers and reflected sound waves in the listening area. The answer given was to get the placement of the speakers to within a quarter inch of perfect then sit still and don’t move. I totally disagree. I move my speakers around by several inches and several degrees without any difference in sound quality. Put your ear next to the speaker. Can you hear the cymbals, yes or no. If no, then the sound of the cymbals never got past the crossover. I would ask the people who suggest the sitting in a perfect spot if they could recommend a brand of vice to put on my recliner to hold my head for me and if snugging the vice would enhance the listening experience.

(Warning: Speaker builders have known that different capacitors sound different since the 1950's. Some people who rely on meters and stats haven't been able to wrap their head around that, and they don't want to listen for themselves. Since they haven't proved it and they haven't listened, they say we are nuts. I say, let your ears be the judge.)

I have been trying all my low volt caps to see which ones work best together. My requirements are simple. A sound stage can be several sound booths, or a small, medium, or large stage. I want to hear the difference from song to song. I want to hear clear vocals, and hear the cymbals. I got the best results from using the

Dayton 1% polypropylene, 250 volt,, and the paper caps. These are the paper caps I recommend.

Gudeman XFS-1952 Axial Paper (CFPA) 001GDCA 1MFD 100V,

Sprague Axial, Fixed Paper 3.5 MFD, 200V, (CFPA) 003R5GADA,

Sprague Axial Paper Dielectric (CFP) 118P60592S4, 6MFD 200V.

Gudeman Axial, P/N: XHF-2535-10 Paper (CFP) 005000GADA .005MFD 200V,

Sprague 186P Axial (CFP) 010000GACA .01MFD 100V,

Gudeman XHFS6270333B Paper 2% (CFP) 033000GBDA .033 200V,

Gudeman XGS 10% (CFP) 100000GAFA .1MFD 400V.

Found here,, The size links are on the top left corner of the page.

(Edit. Using 100v and 250v caps together works well. To my surprise, using 200v and 250v caps together results in cancellation of the finest detail in the music. The finest detail is separate from other sounds in the recording because the wave length is so small, as in razor thin. Using a scope to find them is very hard to do. The time delay of the 50v difference harms the most sensitive signal. The time delay when using 250v or more difference in capacitor working voltages can be detected because it is in the top end of the hearing spectrum. Much of the fine detail, on paper, is at the very top and above the range of hearing but not above the range of perception. The cymbals are still there, but they are weak and they lack luster. The best way to avoid all this mess is to use only one size voltage on all the caps in the crossover. 630v caps are a good choice and a middle of the road sound, halfway between bright and soft.)

(Edit, 2-10-15. Caps make all the difference. I just finished a new 2-way using two 8 inch guitar speakers, 4 ohms each, and a Dayton Planer, PT2C 8 ohm. I crossed the Planer at 9500hz, second order, 1.0 mfd 100 volt Gudeman, and a .268 uH coil, and no crossover for the two 8 inchers. I missed the sound the tinfoil added to the low end so I added a second order to the low end and crossed over at 8500hz. I used two Mundorf ZN 2.2 mfd, 250 volt and a .33 mfd, 250 volt Dayton 1% with a .075 uH coil and all the good sound in the low end came back. The Mundorf ZN and the Dayton poly 1% mix well and the clarity the tinfoil brings made all the difference. I get all the good sound of quality caps and no crossover distortion below 5,000hz. and very little above 5,000hz. The Peavey 8 inch are not made anymore but these,, are about the same thing in 8 ohms.)

Using these caps (1 Dayton and 2 100v papers, or 2 Dayton and 4 100v papers) I get huge depth of sound stage when a large stage was used and the small stages have very good stage imaging. The cymbals are clear and the quiet ones are present. The vocals are clear when the singer has a good voice. When the studio used wave shaping circuits to improve the vocalists performance, the vocals are a bit fuzzy. (Mundorf ZN caps, 100v or 250v, mix well with the Dayton 250v and the Gudeman 100v caps, the Dayton cap should be the smallest of the group.) That brings us back to the newer technology bringing out the noises we couldn’t hear before. I’m listening to my favorite music that I have been listening to since 1955 and am hearing sound that has always been in the recording and is now being played into my living room on a pair of speakers that don’t add anything and don’t omit anything and are free of cancellation.

EDIT: 7-21-15

I’ve recently added several brands of caps to my collection and used them to compare them and see if there were any that worked well with the caps I had. I wanted to find a very high end combination for those with the funds to buy them. I also made tests to see if there were different working voltages that could be used that didn’t cause cancellation. I found that a difference of 300 volts was especially bad for the cymbals. A difference of 500 volts was bad for the mid bass and a difference of 700 volts made the lowest notes fade very distinctly. A difference of 900 volts worked very well. I was surprised but glad because it meant I could mix my 100 volt caps with 1000 volt caps. I tried a difference of 1400 volts thinking I was on to something but all the low notes sounded very strange and muddy, so I stopped using those voltages with each other.

The new cap that worked much better that I expected, when mixed with other caps, was Rel-Cap tin foil-Teflon 100 volt. That was my first 100 volt Rel-Cap. I used it with Mundorf ZN (tin foil) 100 volt and the depth of the sound stage was the best I ever heard. I have used the Rel-Caps before and they were clear and always had good depth but when using those 2 together the depth added to each other. After about an hour of listening to them I could feel some fatiguing (pressure on my ears) from the extra clear sound they have together so I added a Mundorf Silver-Paper-Oil to smooth the sound out. I’ve been listening to that combination, 4.7 mfd Mundorf ZN 100 volt, 0.1 mfd Rel-Cap tin foil-Teflon 100 volt, and a 0.1 mfd Mundorf Silver-Paper-Oil 1000 volt, for a couple weeks now and the fatiguing is gone, all vocals are as clear as they can get, no cancellation and the best depth I have ever heard. These are the most expensive caps I will recommend because they work together better than anything else I have ever tried. They sound so good together that I’m going to stop looking for a “best combination” of caps. I’m going to build one more 2 way and use that combination on the woofers and only use the Mundorf ZN 100 volt caps on the tweeters to save money and still get the most out of my money.

Edit: 11-20-15

I did try a low cost cap called MultiCap and it is better than the old stock paper caps. I tried their Polypropylene Film & Tin Foil Capacitor, 100 volt, and it goes best with the ZN and Rel caps. Those 3 caps sound almost as neutral as a bare wire, and that is as good as it gets. Partsconnexion has them.


Link to pics.

There is a photo of the stretching rack here, click on the pic and then magnify for a closer look.

1, This is what all 4 corners look like. Be sure to have at least 2 of the threaded rods held to each other at the corners.

2, Use a lot of staples to get the stretching even.

3, This is a polyester cloth with a course weave. Cotton is too tightly woven and will block too much sound.

4, Stretch the cloth until it starts to tear at the staples.

5, Stretched cloth, large enough for both speakers. Let sit for at least two days then adjust all 8 nuts (½ X 13 threads) tighter, 2 more full turns and let sit for a couple days.

6, All the gap is on these 2 sides. To get a best fit put the cabinet face down on the board that the covers will be made from and trace the exact fit onto the cover board.

7, If the cover material is very thick, this is about how much gap is needed or the finished cover won’t go in place. If the cover material is very thin use only 1/16 th inch total gap. It is much better to have thin gaps around the covers than to have to fight to get them in place because the gaps were too small.

8, The front cover board is in place and the dark lines are where I will cut out the opening.

9, This is important. Before cutting out the inside, I use a grinder and put chamfers (corner breaks) all the way around the edges. The top inside corner gets extra clearance ground in for the folded cloth to go in so there won’t be 3 layers of cloth sticking up. The grinder burnt the wood a bit, notice the grind does not go all the way to the front of the cover. It goes near the front but does not change the shape as seen from the front. More later

10, Front cover frame cut and painted. The numbers around the frame show where to put the first 10 staples, in order. Once those first 10 are in, go around the frame putting 1 staple half way between the ones that are there, pulling the cloth as tight as you can each time. Once the first 4 are in make sure the cloth is on straight before continuing.

11, This is an extra that is optional. It is for blocking the sun from shining thru the cover and lighting up the drivers. Most people don’t want to see them and bright sunlight shows thru the covers. The thin cloth is not stretched tight. That works better to block the sunlight.

12, Thin cloth in place.

13, Start putting in the first 10 staples.

14, I’m pulling the cloth down as tight as I can.

15, This is what it looks like after going around the outside a few times.

16, This is what the final round looks like. You need this many staples to get a good tight cover.

17, The hard part is done, now to do the corners.

18, This is the corner before you fold the two layers into the ground down corner on the top of the frame.

19, Pull the corner pieces over and back as tight as you can and staple them down, putting the staples into the ground down part on top of the frame.

20, This is one of the other corners stapled down on the bottom of the frame. I do my corners on the tops and bottoms. But that doesn’t matter as long as they go into the ground clearance on the frame so the extra cloth doesn’t stick up and cause big problems.

21, With all the staples done, cut off the extra cloth all the way around and flush with the frame. The extra cloth will just get in the way.

22, I use masking tape to go all the way around the frame and tape down the loose threads. This also stops the cloth from unraveling.

23, This is the finished cover from the inside.

24, This is the finished cover, front side with a 16 oz. hammer laying on it. No sagging, just barley a dent at the hammer head. The cloth is almost as tight as it was on the stretching rack.

25, The pieces stapled on the back of the cover are the same stuff I use for gasket material. One or two layers may be used to get all the edges flush to the cabinet edges. I always make the recess in the cabinet for the frame and cloth just a bit deeper than it needs to be so I can adjust the flush fit of the cover to the cabinet. I use 1- 5/8 inch #6 cabinet screws, drill holes in the baffle board and hold the cover in place with 6 screws from the inside. To take the covers off you have to take the rear board off the cabinet. This stops cats from using the woofers as clawing boards and stops fingers from pushing in the tweeters.

26, These are thin cardboard from the back of a pad of paper used to evenly space the sides of the cover. They are in very tight but do a good job of getting the frame in the right place. The gaps on the sides are too thin to be able to see the masking tape. If you want to hide the gap a bit better use black hokey stick tape on the sides of the cover. One layer should be plenty.

27, The cover is on and flush all the way around.


28, On the back of the cabinet drill a small hole that is part in the rear cover and part in the cabinet.

29, Two drill sizes and one long panel nail.

30, Use the smaller drill and make a hole for the nail, drilling from the inside out. The 2 drilled holes must intersect.

31, Drive the nail into the rear cover, cover the nail and the hole in the back with wood putty. Put one nail on each side of the cabinet. No one will get that nail out without making a mess of things. That way you will be able to tell if someone violated a warranty. If you have to make a repair, use the first drill and re drill the first hole, cutting the nail. You can put a second nail next to the first if you want to.


Damping must be a secret that the big box stores haven’t figured out yet. Or maybe it’s against their religion. Truth is that damping is responsible for a good stereo image and depth of sound stage when a large stage is used to make a recording. Without proper damping the sound waves inside the enclosure bounce around and then leave the enclosure by going thru the paper cone of the driver. That causes pressure on the cone that is out of phase with the motion from the voice coil and that destroys the pristine reproduction of the original recording. It’s like standing on stage and trying to sing while someone is smacking you on the back of the head with a tennis racquet. And of course there is the problem with cabinet resonance. Cabinet resonance is like an air horn that will play itself every chance it gets. Both problems have got to go.

Fiberglass almost works. It was used for a long time when there wasn’t much else to use. Nowadays there are places that specialize in sound absorption. This is a material that has been given the highest sound absorption rating the industry has to offer. This is probably better but I have never tried it. I’ve thought of using both, the cotton on the side walls and the Elves Hair for filling the enclosure. Maybe I’ll try it some day.

There are 2 resonant sounds in an enclosure that need to be removed. First is the sound of the wood walls themselves. To do this staple the cotton tightly to the walls and baffle board as seen in the first pic,, It has to be tightly stapled so any sound pressure has a dense layer to go thru to get to the wood, and a dense layer to return thru to the chamber. If there is no covering over the walls and you hit one side with a hammer it would make a lot of noise, cover the walls, hit one with a hammer and there will be almost no noise. Sound waves will act the same as the hammer.

The second resonant sound is the chamber itself. Tear the acoustic cotton into small chunks. Maybe even tug at them to loosen them up a bit. Then use them to fill the chamber but don’t pack them down at all, not even a little. You want them loose and fluffy. To prevent the loose material from getting to the drivers place a thin porous cloth, not stretched very tight, over the rear of the driver and staple it down. In the photo I am using 2 drivers with open backs so I’m covering both with the cloth. If I’m using only 1 driver, and it has a low profile, I will make a wire cage to go over it to hold the cloth away from the driver and have some room for the driver to breath. This pic shows the coat hanger wire used to make a basket behind the driver to keep out the loose stuff.

The last pic in the series shows the enclosure full of shredded ‘blue stuff’. It is very loose and will stop any cabinet chamber resonance dead in its tracks. It will also stop sound from bouncing around and interfering with the motion of the cone.

I’m sure there are better ways to damp an enclosure but this works so well that I stopped looking for anything better. My cabinets reproduce life like stereo imagery and depth so deep it extends thru the living room wall and into the kitchen. Hollywood has been recording sound with light beams since 1930. The recording industry didn’t change to laser beams until about 50 years later. Soundtrack music sounds GREAT because Hollywood knows how to get it right. I play the audio from my DVD player with my speakers and they make surround sound a waste of time.

Damping the dust cover, and sometimes the cone, is also important. See "In the absence of resonance.....". Damping the resonant sounds of the drivers is equal to damping an enclosure. It makes that much of a difference. I'm sure the + $10K speaker systems use drivers that have had their resonant sounds eliminated.


Edit 1-25-16

FaitalPRO 10FE200 10"

FaitalPro midrange, M5N8-80,

Or the Pyle PDMR9 midrange. The Pyle has been discontinued for some time but they keep showing up. It was one of the best sellers they ever had. This Pyle is an 8 ohm. They say 4-8 but I don't think the suit that wrote that knew what he was talking about.

Tymphany (Vifa) XT25TG30-04 1" Dual Ring Radiator Tweeter

I built this 3-way using the Faital 10", PDMR9 and the XT25TG30-04. I don't use a crossover for the woofer. The crossovers used are at 800hz and 4500hz. Be sure to wire the mid and tweeter in reverse to match the phase of the woofer.

The parts list and projects that were on this page have been removed because the midrange driver I was using is no longer being manufactured. That has been a longstanding problem with speaker building. It seems that Faital is the only driver manufacturer that is willing to keep selling proven drivers. If you plan to build for profit I suggest using the Faital products simply because they will be around a long time. Parts Express, Amazon, US Speaker and others sell their drivers.

I found this 5" midrange driver, M5N8-80, at Faital,, that looks good as a midrange. The M5N8-80 is costly but looks great. I really like the polycotton surround. All the best drivers use cloth for the surround. The other two 5" drivers use a different paper cone. Make notice of that. The dotted paper cones that I have used in the past were all a disappointment because they had some kind of poly impregnation that was brittle sounding. The paper used in the M5N8-80 and the 10FE200 are the same. That means their tone will match each other.

The M5N8-80 has a listed VAS of only .01 cubic feet, or only 17.28 cubic inches. It looks like the magnet of the driver has about 25 cubic inches to it and that must be added to the inside cubic dimensions of the little box you have to build to make it a sealed back enclosed driver or the pressure from the woofer will distort the midrange. If their numbers are right then a small box with inside dimensions of 4" X 4 " X 2.75" will be large enough for the VAS requirement, when the mid is mounted on the front. Be sure to use damping material inside the small box, the same with the rest of the enclosure. See my post on damping. The 99 SPL is quite high and will need a series resistor to lower its loudness to match the rest of the system. See this chart: and this chart:, for the SPL corrections. This page is for the crossover calculators:, The Bessel style crossover has a bit better sound but needs larger capacitors. That can raise the cost a lot if you are using the very expensive capacitors.

With a 15" woofer I would crossover between 800 hz and 1100 hz. With a 12" woofer I would crossover between 800 hz and 1400 hz. The wiggle room is for capacitor availability. I would crossover the tweeter between 5 khz and 6 khz. Both crossovers would use 1 coil and 1 small group of capacitors, all with the same voltage rating. I wouldn't use a crossover on the woofer, it isn't needed. Don't worry, the electrons will know what to do. Reverse the polarity on the mid and tweeter to match the phase of the woofer.

There are enough tweeters being made these days that if a favorite goes away it won't be hard to find a different one.


This a Vifa tweeter with a skinny POSITIVE terminal. ALL the other drivers I have seen use the wide terminal for the positive.

I destroyed one of their terminals a while ago because I wasn’t expecting the plastic holding the terminal to melt so fast. I was wrong. Now I’m more careful. The series of photos,, shows what to do.

The first pic shows the skinny piece standing up from the wider piece that goes into the thin plastic holding pieces. Use a screwdriver to bend the skinny part to about a 45 degree angle. Put a rubber band around some needle nose pliers, then grip the bottom of the skinny piece with the pliers, the rubber band will hold the pliers in place so the heat from the soldering will stop at the pliers and not get near the plastic. Strip the wire you are going to use and put a piece of heat shrink over the wire. Place the wire into the hole, solder it, it will be crooked so use the pliers to straighten the wire and slide the heat shrink over the solder joint and use the soldering gun to heat the shrink tube just enough to form fit to the solder joint. Find some tape with strong adhesive and marry the tube to the tweeter body for maximum strength and strain relief. (that’s not duct tape, it’s aluminum foil tape, shipping tape with the strings in it is also good)


Capacitors are not all created equal, therefore they do not sound equal. Links to different sound characteristics and evaluations are below. Please note that the vocabulary from one person to another is quite different. There are no carved in stone definitions of what capacitors sound like. But they absolutely do not sound the same. Music is far to complex to not be effected by the many small changes in manufacturing materials and process. Not much has been done to compare caps one to another because of the high cost of the caps and the long time it takes to do a good job. All of the testers listed below used different amps and speakers, mostly high end, (I think most of the testers are rich kids and/or are addicted to finding the best way to play their music).

If you want to learn about the confusing world of audio capacitors you have to start somewhere. These links are a good start. Please keep in mind that capacitors in non audio applications sound different too, except no one is listening or cares.

This first link gives a chemists view of why capacitors sound different.

On Capacitor Dielectric Materials - A Chemist's View Karl A. Weber, Ph.D. (long winded, but he knows his stuff). He talks about dielectric absorption, or energy loss due to the type of dielectric used in the cap. The razor thin fine detail of sharp notes are the first to suffer losses to DA. He says paper and oil are hard to measure, says nothing about dry paper. His findings are that Teflon and polypropylene have the least absorption of energy, polyethylene and polystyrene have more absorption. I expect he avoids dry paper caps because they are obsolete, however, they don't seem to absorb the fine detail and were used extensively because they sounded great and were dirt cheap.

“For over twenty years, audiophiles and audio designers have been aware that not all caps sound alike. Capacitors can make a huge difference in the sound of equipment.”

“The quality of the capacitors, specially when they are in the signal-line, have a very big impact on the sound-quality of an audio-circuit.”

“Actually, no one likes comparing capacitors. Do we really have nothing better to do in life but to swap capacitors in an out of circuit, burning in numerous capacitors just to find out how they sound ?”

“many capacitors today use an insulating plastic film as the dielectric because it makes them inexpensive to produce, but the problem with using plastic is that it noticeably degrades the audio signal, “With plastic you get what Steen would describe as an echo, masking the real details of the signal”.

What Audyn says about their True Copper caps.,en,6559

Audyn True Copper Cap

A very high end capacitor should be absolutely neutral and should pass signals undistorted. The Audyn True Copper Cap does not have a seperate sound or sound picture. Absolutely exact, accurate and honest reproduction of the signals that enter. This capacitor is not an Equalizier, not pointed, not emphazised, not warmer, not fresher, not brighter, not deeper, not wider. It simply does what it should. Do not buy, if your audio devices are not of high quality.

Full OFC copper-foil polypropylene capacitor for exacting and high quality audio applications. No contamination by strange mixture of various metals and fairylike oils.

It is True Copper

This is what Auricap says about Auricap.

“With the design changes we have made a new level of performance is reached with even less coloration, micro-phonics and noise contribution than the famous Auricap.”

This is what Jupitor says about Jupitor.

“Jupiter copper foil wax & paper capacitors go far beyond a mere copper foil version of the standard tubular HT Aluminum design. Every aspect of the capacitor construction was evaluated during the 4 year development process. Even the smallest details were addressed which resulted in significant improvements in sonic performance. Wax with very special sonic characteristics, better damping and thermal characteristics was chosen as a perfect match. Further, new lead termination technique has allowed the music to flow more naturally. Without exception, every detail has been tailored to this specific copper capacitor design. After four years of refinement Jupiter Copper Foil capacitors are here. Life like vibrance gives the feeling of presence, along with micro-detail, holographic bass, and aural decay that is disarmingly natural, the ears will know immediately.”

Robert Hovland of Hovland capacitors says:

"All capacitors, no matter who makes them, have issues with consistent sonic signatures throughout the voltage ratings. This is because of the different dielectric thickness requirements (basically determines breakdown voltage, all other factors equal). Lower voltage caps will always have a tendency to be faster sounding and more detailed, the higher voltage caps will always have a tendency to be more laid back. In good systems the pacing of the music will also be faster and slower respectively."

The "plate of glass' theory comes to mind. A single plate has no visual distortion, but enough plates, stacked up, will blur the image. You will still see the basic outline of what you are looking at but the fine detail will be lost. 1200 volt caps have too many plates of glass for the razor thin fine detail to get thru clearly, if at all.

EDIT: 9-10-14

I've often heard of fast caps but couldn't reproduce what I was told about them until I got rid of the dust cap and cone resonant sounds the driver manufacturers built into their product. See "In the absence of resonance....". Now I can hear the difference between the 200V caps and the 1200V caps. The low volt caps are sharper and more detailed, most 1200V caps I tried are much softer around the edges. The Jantzen Z Superior is soft but nicely warm. It would be nice if the cap makers made more 200V caps available. The Sprague and Gudeman sealed paper 200V caps ( are a bargain compared to other big name 200V caps. The Mundorf ZN 2.2MFD caps are 250V, the smaller ones are 630V. Mundorf has added 2.7MFD-4.7MFD 100 volt to their ZN line. There seems to be a difference in sound every 200V. The 630V caps are in the middle of fast and slow. The author of Humble Homemade says the high volt caps always sound better. I disagree. He must like the soft sound more than I do, I like the crisp detail and the low volt caps do that better. I've spent the last month mixing caps by voltage to compare and the volts make a bigger difference than the dialectic absorption, the DA is mostly a smaller loss. Using a 2 way, leave the woofer cap alone and change the tweeter cap to different voltages and hear the difference. I'm sure there are some dialectics that ruin the fine detail when used with 1200V ratings, but neither are as bad as the resonant characteristics of the dust cover or cone. Damping those sounds brings a change to the sound envelope equal to damping a hollow enclosure.

EDIT: 11-11-14

I've done more testing caps by voltage. I made a chart showing several caps and their voltages, Mundorf MKP and Erse Pulse X both make several caps from 100v to 1200v, but didn't have room for them in the chart.

When using caps with a 50v-100v difference you get cancellation of the highest frequencies. These are mostly harmonics and detail. The loss smooths the sound and the depth of image goes down a little bit. A difference of 150v produces a loss that isn't as noticeable because the depth of image doesn't lose its detail and not as many instruments have harmonics in the range where the losses occur. This chart shows the instruments range below their harmonics. When using 2 caps with different voltages the cancellation will be maximized. When using 4 caps and 3 have the same voltage rating the cancellation will be reduced proportionately. Differences of 200v and higher will put the losses in the upper ranges of the instruments sound. When using many caps, each with a different voltage, the cancellations are spread out and may even produce a smoother response curve at the drivers construction materials resonance and the resonance of the dust cap chamber, by accident, and it might even help a little. Maybe. Using only one voltage of capacitor is best.

A big problem with high voltage caps is that the MFD's may be 3% but the voltage could be +10% or more. They don't tell us. Using 2 or more in parallel could be a source of delay even if the caps cost a fortune. The 100v caps can't do that. Even at 15% tolerance on the voltage the frequency of cancellation will be way above the range of hearing.

All the over 800v caps have a dielectric thick enough to remove (impede) a small amount of detail in a way that improves musicality. The energy of the small detail is overcome by the higher impedance of the thicker dielectric. The 100v caps don't impede anything. 630v caps are for the people who can't make up their mind or want the best of both worlds.

I put a few 100v caps, ZN and Gudeman, in parallel and put them in series to a 2.2MFD 900v cap, put a 1000v Mundorf Silver/Oil across them and put four 0.47 250v Dayton 1% in series to get a 0.117MFD 1000v cap and put that across the others as well. I also put a 6.0MFD 200v Sprague paper and a 6.8MFD 800v Jantzen Silver in series to each other and in parallel to the others. They took a couple days to settle down but sound really good. The signal saw 1000v with no delayed signal and I can hear the sound each brand has. This mix gives rich, full sound and only a little loss of depth of sound and detail, but still my favorite mix is one 100v Mundorf ZN with one 100v Gudeman Paper and one 100v Rel Cap Tin Foil/Teflon with one 0.1 MFD 250v Dayton 1% bypass cap for best detail and best depth of image. The Dayton cap took the edge off to smooth things out and reinforce the depth of image. A couple of weeks after I started using that mix, the caps smoothed out and settled down and lost their extra sharp edginess. The cancellation from the Daytons showed up a bit so I took them out and was surprised how much of an improvement was made. Using only one voltage of caps is by far the better way to go and using 2 or more brands in parallel sounds better than only one cap per circuit.


That’s the name I’ve given to corporate made speakers. Stockholders decide to make some money and they think speakers are a way to do that. They go about the business of putting parts in a box and then put the box in a big box store. They hire a marketing PHD to invent a cute sales line for the salesman to recite.

Corporations are in the game to enrich themselves as best they can and customers are a necessary inconvenience. I say this because I have bought well over 50 pairs of blown speakers at garage sales over the last 40 years so I can take them apart and see what went wrong. Usually, the spiffier the outside of the speaker, the cheaper the inside looks. As in real cheap. I take a forensic look at the crossover to see what is really going on and I can’t believe my eyes. Plenty of ’make money’ logic and no speaker theory logic at all. I got my electronics diploma from Brown Institute. I can’t imagine where they got theirs.

I can’t count how many times I’ve seen a crossover point 2 octaves above the beam frequency of a driver. Drivers are like a flashlight that has a twist head so the light can be forced into a beam. Every driver has a beam frequency. It is = to the speed of sound in inches (13,740 inches per second) divided by the diameter of the cone. A 4 inch cone will start to force the sound into a beam at 3435hz. That beam will be almost fully closed at about 3800hz. When the sound is in a beam it can’t form a ’sound stage’ sound. The sound has to spread out to do that, and a beam isn’t spread out. I’ve seen crossovers that will let the woofer go into a beam, and both the mid and tweeter only play sound above their beam frequency. What a joke.

Speakers are supposed to do more than just emit sound. They are supposed to recreate a sound stage presentation, when the music was recorded on a sound stage with microphones. I think the closing of stereo stores over the last 40 years is a blessing in disguise. It allows for the building of your own or have a friend do it for you. Then you will have personal satisfaction vehicles.


I have several brands of caps to use to see how they match up to the Titans.
I have:

Mundorf Supreme

Mundorf ZN

Mundorf Silver/Oil

Mundorf Silver/Gold/Oil


Intertechnik, True Copper

Duelund, Alexander

Jantzen Z Superior

Jantzen Silver

Sprague Paper

Gudeman Paper

And a few others that I don’t use anymore, some because they don‘t make them anymore.

All of the big name caps do change the color of the sound but for the most part the change is a step sideways not up. There were 2 that made an improvement that could be heard, they are the Jantzen Silver and the Mundorf ZN. I used 0.1MFD. My crossover coil sizes were made for a 3.3MFD cap and the Titans are 3MFD +/- 5%. I have 10 of the 3MFD Titans and checked all of them and they were all either 3.08 or 3.09MFD. When I tried both the Jantzen Silver and the Mundorf ZN across the Titans at the same time, the improvement was quite impressive. Those 3 caps like each other (see audio capacitor characteristics 1 & 2 in the side bar). All the caps I used were already broke in and I let them play for a few hours to settle in anyway. Concert hall sound, depth, sharp and clear vocals, excellent cymbals and percussion. Plenty of detail. Not harsh or overdone at all.

The Titans, by themselves, are quite good, and cheap. They are made for motors and air conditioners so they can use the help from a small bypass cap or two. They are not intended to compete with $100.00 per MFD audio caps but they will have the big boys looking over their shoulder for a long time to come.

I've tried many combinations of caps but keep coming back to these low cost oil filled caps for their smooth and deep sound stage reproduction. The paper 200V Sprague or Gudeman ( caps, from 6MFD down to .005MFD, sound great with the Titans, and the Mundorf ZN caps, even the 650V ones. The Sprague and Gudeman are a step up from the Jantzen Silver, mostly because of the "low volt higher speed". See "Capacitors really do sound different."

1 comment

SPL= Sound Pressure Level. It means the efficiency of a driver. Matching the efficiency of the several drivers is very important. That way the loudness of the sound is the same from each driver in the speaker enclosure. Avoid using drivers that don’t have a SPL rating. The SPL rating system is for a 8 ohm driver. If you are looking at a 4 ohm driver the SPL rating is still given as a 8 ohm equivalent driver. If the rating says 93db, you will get 93db if you put a 4 ohm resistor is series with the 4 ohm driver, to make 8 ohms. If you don’t use the extra resistor the 4 ohm driver will draw twice the amps and you will get twice the power and that is = to a 3db increase of loudness level. That means a 4 ohm driver, rated at 93db, can give you either 96db or 93db. it's your choice to make the 4 ohm driver like a 8 ohm, with a resistor, or not. A 8 ohm driver rated at 93db will give you 93db of Sound Pressure Level.

If you have a 92db SPL, a 95db SPL and a 98db SPL driver, all 8 ohm, you will need to reduce the 98db and 95db drivers down to 92db's each with either a L-pad or a equivalent series resistor to compensate. This will make the outputs of the 3 drivers the same loudness level.

If you don’t know the SPL of your drivers then you are in for a hard time. Adjusting the SPL by adjusting the resistance of the circuit is the way to compensate for different SPL rated drivers. If the resistor is placed between the driver and the crossover, you have to add the resistors value to the drivers ohms to calculate the proper values for the coil and capacitor in that circuit, (woofer, mid, or tweeter). If you put the resistor before the crossover parts for that driver, you won’t have to change the drivers ohms rating to calculate the coil and capacitor.

See the sidebar for charts giving the ohms values for a single compensation resistor to be used in series, or the two resistors needed for a L-pad to lower the drivers output without changing the total resistance of that circuit.

Crossover, L-Pad & Zobel calculators Most builders use the second order, Linkwist-Riley values. Bessel is better but the capacitors cost more because they are a bit larger.

Not all speaker manufacturers use the same method to measure SPL, or they don't get it right. Sometimes I get a driver that is much louder than the rating says it will be. It happens. Just use a larger resistor to quiet it down. An average pair of headphones make a nice reference for comparing loudness levels of your drivers. If a driver is too loud or quiet, the headphones will let you know.


If you were having a party in a 2 car garage with 20 or so people, and you were hearing 10 watts of music, everyone would have to talk as loud as they could to be heard by a person a few inches away from them. If you wanted to turn the stereo up so it sounded twice as loud, you would have to turn it up to 100 watts. If you wanted to turn the stereo down so it sounded half as loud as 10 watts, you would have to turn it down to 1 watt. See this chart for loudness comparisons

The reason for this is because the ear has a built in safety feature. The ear is made to protect itself from loud noise. If the ear did not have that protection, a close thunder clap could destroy our hearing and put us down on the ground in pain and we might not be able to hear ever again. Wind blowing at 50mph would sound too loud to be safe for the ear. The ear almost always requires sound to be 10 times louder to sound twice as loud, or 10 times quieter to sound half as loud. This means a 1000 watt amp will only sound twice as loud as a 100 watt amp, 100 watts sounds twice as loud as 10 watts, 10 watts sounds twice as loud as 1 watt, and 1 watt sounds twice as loud as 0.1 watt. Most people listen to less than 1/2 watt. Many of the most expensive stereo systems only go to 30 watts, that’s all. Big box store sales people are told to sell power because most people don’t understand power at all.


The problem is that the spec sheet for testing for SPL is worded so that the sellers of drivers don’t have to tell you that there is an impedance gain left out of the reported SPL for a driver that is less than 8 ohms. A 4 ohm driver will have a 3db gain because the 4 ohms will draw twice the amps and thus twice the power. A 4 ohm driver rated at 94db SPL will actually be 97db SPL. You have to know that yourself, they won’t tell you that. See this chart for cross references. To keep the crossover parts the same size, see this page and put 4 ohms at R1 to get rid of the extra loudness. Or add 4 ohms at R4, to the 4 ohm driver, and use the crossover parts for an 8 ohm driver. The caps will be smaller and the coil will be bigger.

Edit: 8-9-16. Another part of the problem is that not all manufacturers measure the SPL the same way. Some just plain screw it up and get it wrong. I used to always use 6db as the impedance gain for a 4 ohm driver and sometimes that was right, sometimes it was 3 db high, and sometimes it was all over the map. The trick is to listen to the drivers at low power to see if the output levels sound the same at the crossover frequency. Then quiet down the loud one with a resistor as needed. The art of speaker building is the art of problem solving.


Bass reflex was the only new technology for a loudspeaker back in 1930 when people had 3 watt amps, and AM radio with 30db signal to noise ratios. AM only broadcasts down to 120hz. FM broadcasts down to 80HZ. It was hard to find the resonant problems with bass reflex when there was no true sub bass at that time. By the sixties the bass reflex lived on in folk lore and kept alive by parts sellers who knew it was a failed technology.

Acoustic suspension was the next logical move. Shrink the enclosure down from a fridge sized monster to less than 3 cubic feet. It worked so well that everybody started to do it. Many of those models had good woofers and some did not. The difference was the stiffness of the cone. Thin paper cones only lasted about 5 years and then they softened because of the sound energy passing through the cone from inside the enclosure to outside the enclosure. (The energy inside the enclosure is = to the energy outside the enclosure). The high pressure inside the enclosure will find a way out if it can and a paper thin cone is an easy way for much of that energy to escape. The cones that were thick and stiff can last over 50 years. Today’s carbon fiber should last even longer.

After tech school I tried my best to build a good bass reflex system. The problems with resonance are insurmountable. I concluded that resonance was the enemy of a good speaker and resonance must be eliminated, not tuned. That means using good dampening material.

I liked the idea of a vent to let energy out of the enclosure and came up with a way that does not add a resonance to the output. The books say don’t put any material in the mouth of a port or vent because it will destroy the tuned aspect of the port/vent. Good. Destroy it. I use a slit, made with a radial arm saw, in the rear cover board, not the baffle. I clamp the board onto my radial arm saw and lower the blade where I want the slit to be. The outside of the rear cover is face down. I cover the inside of the slit with many layers of cloth to impede the air flow and destroy resonance. When properly done I can listen to what comes out the slit and it does not have recognizable sound. I don’t hear much of anything, even at high volume but I do feel pressure leaving the enclosure. I want pressure to come out, not sound. If the bass is a bit boomy, add more layers or use thicker cloth. At very high volume the pressure leaving out the slit should only move a birthday candle flame no more than the width of the flame when held next to the slit. At lower volume there should be almost no pressure leaving out the slit. This method improves the bottom end of the bass, maintains the punch of the bass when present, and relieves pressure from the enclosure when the pressure is high. I stole the idea from the zener diode.


You need at least a saber saw and a drill, a 3 amp variable speed reversing is good enough. Six amps will break screws. A radial arm saw is better, or know someone who has those things. Then you have to decide how much is too much money to spend and how much space you can use to set them in.
A well chosen woofer and tweeter will make a very good 2 way speaker. They are much less complicated and simplicity goes a long way. An 8” woofer is both small enough and large enough for great sound. The MCM 55-2960 is much better than the $17.00 low cost implies. The Usher 8945A would be my next choice. It is better and cost $129.00.

Read the “Tweeter Mishmash” posting. It makes clear that some “cheap’ tweeters are as nice or even much better than others costing over 5 times as much. With a 2 way it is more important to use a tweeter with a low resonant frequency. The Vifa XT25TG30-04 is listed in the “Mishmash” as one of the best and I use them for their clean sound and low resonance. In a 2 way with an 8” woofer I would cross over at 2100hz.

A 3 way would use a smaller diameter midrange. 12” woofers are common in a 3 way although I think the 10” is quite nice as well. I like the 800hz to 850hz low frequency crossover point for its great natural sound in the vocals. Don't use a band pass circuit for the midrange. It only complicates things. Use a second order c/o and reverse the polarity of the woofer to correct the phase difference between the woofer and mid and tweeter. The mid and tweeter will be in perfect phase and the higher notes will bypass the mid so there will be no impedance gain in the upper frequencies. Cross over the tweeter 3-4 octaves above the tweeters resonant frequency, (fo) or you will get noise out of the tweeter at its fo.

This is a starting point. Read the links in the sidebar, use the links in the “Supplies” post to shop around.

1 comment
Community Details





For loudspeaker builders

Create Post
Cookies help us deliver our Services. By using our Services or clicking I agree, you agree to our use of cookies. Learn More.