All things speaker resistor: questions and answers

[DavidR]

A lot has been written about replacement of speaker xover resistors including adding resistors to film caps to more closely match the ESR of the original NPE caps. I thought it would be nice to have all the questions and answers to resistors in one thread.

 

Some of my questions include:

(1) Unless there is a visual observation of overheating or a cracked/broken resistor, does it need to be changed? (referring to the AR sandcast resistors mostly)

 

(2) Are Mills, Vishay Mills and Mundorf MOX and M-Resist resistors that much better (for SQ) than the sandcast resistors and it is worth the price to replace them 'willy-nilly'?

 

(3) The AR resistors I've encountered are 22 watts. Can you get by with 20 watt rated resistors?; If you parallel two 12 watt resistors to get 24 (or use a 25 watt wire wound) does this affect the impedance, or is it always better to have more power handling ability.

 

(4) ESR adjusting for film caps:

Re: adding resistor to a film cap when replacing NPEs

 

>Ken Kantor (copy and paste from a thread on AK)

"For NPE's older than 25years, try 1 Ohm.

For NPE's over 100uF, and/or between 15 and 25 year old, use 0.5 Ohm.

For contemporary NPE's, use 0.2 Ohm.

There were garden variety, decent, no-name industrial caps, purchased maybe 5 years ago. I tested them right out of the parts drawer, new, no bias or (groan) break in. Everything above about 33uF tested well under 0.2 Ohms. I would use 0.1 resistors, or nothing, in the rare event that these larger NPE's were being replaced by film. (The largest film cap I have around is 16uF, and it tested a bit below 0.01 Ohms.)"

 

>RoyC (copy and paste from a thread here on CSP)

"The ESR readings you measured were similar to John's as well, and was the motivation for his recommending the addition of 0.3+/- ohm series resistors to replacement caps for those who may want to try to mimic the higher ESR of the original caps. (I personally don't ascribe to this practice, as we really don't know what the original ESR was...and the level controls typically show variations of greater than .3 ohms.)"

 

I seem to remember a thread here on CSP where (I think; I'm unable to locate the post) Ken agreed with RoyC that a good all-around resistor value to use would be a 0.3-ish ohm resistor. I understand the placement for this 'added' resistor is in series with the driver and after the cap(s).

[TimmyTonga]

Hi David,

For what it's worth, I recently measured the ESR of every single Sprague cap I ever took out of an AR (I kept them all), they measured either 0.03 or 0.04 ohms regardless of capacitance value, with one exception, a 20 ohm can which read 0.14. These would all date from the mid-seventies. They may have all drifted into this range over time, but the stated capacitance of the cap doesn't seem to have much bearing on the matter.

regards/Tim

 

[Jeff_C]

I do not believe different brands of resistors / capacitors alter SQ. It is only differences in their combined electrical measurements which can do that. Of course it makes sense to try and use brands which are reliable and will measure up well within their stated tolerances after the passage of time.

[DavidR]

Here's a link to a website regarding resistors in series and parallel.

http://electronics.stackexchange.com/questions/78610/how-should-i-connect-four-resistors-to-double-the-capacity-with-the-same-resisto

[iso]

First I should state that I do hear difference between some loudspeaker cables and absolute phase...

But... I think it is a bit silly to pay premium for film cap and downgrade ESR of better and more expensive cap to level of NP electrolytic by adding series resistor. When using series cap in crossover like AR3a, there is far more shift in operating point due tolerance and position of 15 ohm pot than ESR of the cap including proposed series resistor. Which both are in series to signal and tweeter, unless pot is in wide open position. Sound may be better when using film cap for several reasons, but if this ESR issue bothers you why not use NPE:s as they are quite easily available an correct for this period of speakers.

There is nothing wrong with wire wound-resistors in audio band... even ARC used 300K ohm special order Dale ww-resistor as anode load in SP6C phono section. 

Kimmo

[RoyC]

On 11/9/2016 at 3:20 PM, DavidR said:

A lot has been written about replacement of speaker xover resistors including adding resistors to film caps to more closely match the ESR of the original NPE caps. I thought it would be nice to have all the questions and answers to resistors in one thread.

 

Some of my questions include:

(1) Unless there is a visual observation of overheating or a cracked/broken resistor, does it need to be changed? (referring to the AR sandcast resistors mostly)

 

(2) Are Mills, Vishay Mills and Mundorf MOX and M-Resist resistors that much better (for SQ) than the sandcast resistors and it is worth the price to replace them 'will-nilly'?

 

(3) The AR resistors I've encountered are 22 watts. Can you get by with 20 watt rated resistors?; If you parallel two 12 watt resistors to get 24 (or use a 25 watt wire wound) does this affect the impedance, or is it always better to have more power handling ability.

 

(4) ESR adjusting for film caps:

Re: adding resistor to a film cap when replacing NPEs

 

Hi David,

My answers are:

1) No...but you should always measure them if there is any doubt. The New Large Advent is notorious for having a single bad 1 ohm resistor. Visually it looks OK, but doesn't pass the signal.

2) No reason to purchase fancy pants resistors, especially given much more significant variables in 40+/- year old speakers.

3) Yes, 20 watt resistors are OK.

-Using parallel resistors to achieve a desired value is a perfectly acceptable way to increase power handling with no negative sonic impact.

-Later AR-3a's were equipped with a sand cast 5 watt .5 ohm resistor in place of the earlier nichrome wire resistor in the mid circuit. Interestingly, I've seen melted "20 watt" midrange pots in the same circuit as healthy 5 watt resistors.

4) After doing some research on differences between capacitance meters and associated differences between measurement results for npe vs film caps, I'm increasingly becoming an advocate of using npe's to replace npe's when the goal is to retain "original" performance. (Ken Kantor posted this view rather emphatically some time ago in another forum.) ESR and capacitance varies with frequency, and a single fixed resistor is not going to compensate for all the behavior differences between an npe and a film cap.

Roy

[owlsplace]

2 hours ago, RoyC said:

I'm increasingly becoming an advocate of using npe's to replace npe's when the goal is to retain "original" performance.

Perhaps with a caveat regarding Classic era speakers that started out with foil and ended with electrolytic caps.

Roger

[ra.ra]

26 minutes ago, owlsplace said:

Classic era speakers that started out with foil

Just curious... which Classic era AR's employed foil caps? I don't think I've ever seen this in any speakers I've encountered.

[RoyC]

I'm not aware of foil caps being used in AR speakers. Roger may be referring to the oil caps of the 50's and early 60's? They are unique animals...

Roy

[owlsplace]

Nope, I was referring to the ICC wax caps as foil caps, which they are are if you have ever taken one apart

Roger

[RoyC]

Those aren't foil caps, Roger. They are oil impregnated paper...different.

Roy

[genek]

If you rip those old paper caps apart, you'll find foil interleaved with the paper.

Capacitors are categorized by the substance used as their dielectric, i.e., paper, generic film (usually polyster), a branded film (Mylar, Teflon, etc), etc. "Foil cap" is a mostly meaningless term, except when used to differentiate between a cap that has foil interleaved with its dielectric vs one where the dielectric has a conductive layer deposited directly on its surface ("metallized").

 

[owlsplace]

Some say "toe-MAY-toe" some say toe-MOT-o, the premise for the original argument remains the same. For example, AR designed the AR-5 with one style capacitor and finished production with a different style capacitor of the same value. Did the two different versions of this model sound different because of the different style capacitors? If so, in what way?

It follows that replacing the original caps with a different style cap, such as paper with foil or electrolytic, may change the sound of the speakers. I don't think anyone is going to go there at this point -- not that I couldn't since I have foil and electrolytic capped AR-5's available.

The later models designed with electrolytic capacitors from the start would be a different consideration; however, regardless of the capacitor type, capacitive reactance still varies inversely with frequency. So, if there are audible differences between the capacitor types it would have to be caused by non-linear losses in the audible frequency range relative to the differences of design.

Got to love those simple Classic era designs

Roger

[DavidR]

19 hours ago, RoyC said:

Hi David,

My answers are: .........................................................................................................................

4) After doing some research on differences between capacitance meters and associated differences between measurement results for npe vs film caps, I'm increasingly becoming an advocate of using npe's to replace npe's when the goal is to retain "original" performance. (Ken Kantor posted this view rather emphatically some time ago in another forum.) ESR and capacitance varies with frequency, and a single fixed resistor is not going to compensate for all the behavior differences between an npe and a film cap.

Roy

Thanks for the responses.

So for NPE caps as frequency increases measured capacitance goes down and as frequency decreases capacitance goes up. Is this not true for film or do they swing to a lesser degree or not at all.?

Would these be any good for replacements on the level switches?

 

[ar_pro]

I am currently using the Erse resistors in my AR-3a/AR-91 hybrid; their value is right on the money, which is more than could be said for the original AR-91 resistors.

[owlsplace]

I decided to try some Mundorf's in the current AR-58s makeover mainly because the idea of putting resistance in series with voice coils doesn't appeal to me.

 

 

[RoyC]

10 minutes ago, owlsplace said:

I decided to try some Mundorf's in the current AR-58s makeover mainly because the idea of putting resistance in series with voice coils doesn't appeal to me.

 

 

Roger,

Sand cast resistors like the Erse resistor mentioned above work very well, and are integral to most crossover designs. What do you hope to gain with an expensive "resistor"?

Roy

[RoyC]

2 hours ago, DavidR said:

So for NPE caps as frequency increases measured capacitance goes down and as frequency decreases capacitance goes up. Is this not true for film or do they swing to a lesser degree or not at all.?

David,

Crossover film caps maintain "rated" capacitance over a wider frequency range. Capacitor specs are usually stated as values at 1000hz. Many meters measure anything over 20uf at 80hz or less, and a film cap will measure much more closely to its rated value than an npe will at that frequency. At 1000hz, however, the npe and the film cap of the same rated value will be much more closely in agreement. Does this cause significant sonic differences? Who knows...but I suspect it could be greater than an overpriced series resistor (sorry, couldn't "resist", Roger. )

These things are always entertaining as topics of speculation and debate, but are probably insignificant given the original, and age-acquired, variations inherent to these old speaker systems.

Roy

[owlsplace]

15 minutes ago, RoyC said:

... Does this cause significant sonic differences? Who knows...but I suspect it could be greater than an overpriced series resistor (sorry, couldn't "resist", Roger. )

These things are always entertaining as topics of speculation and debate, but are probably insignificant given the original, and age-acquired, variations inherent to these old speaker systems.

Roy

"Old speaker systems" -- LOL -- that would be us since the observer is always the primary factor in any experiment!

The gain in using those German resistors would be similar to the gain observed after riding a Harley-Davidson all day compared to riding a BMW motorcycle which is to say you would still be able to walk when you got off

In all fairness though, not too many of us sit around listening to music all day.

Since we are on the topic of series resistance, AR used this technique to balance driver output which also seems to have led to their vacillation over how to phase individual drivers.

Perhaps veering away from the original topic though and the premise for this forum's existence. At any rate I'm feeling insecure about my ability to test this thesis at the moment.

Roger

[DavidR]

Thanks Roy. I got it. I wish I was able to change the frequency used when measuring caps. I do remember having issues with Erse NPEs; however, I did find Bennics to measure much closer to stated values. The Erse would actually be outside of the tolerence (on the high side).That certainly makes xover restorations more difficult.

The reason for the resistor question was mainly due to an issue with my 90s. The amp I had prior to the one I'm using now was a bit forward in the upper mid and at times 'bright' in the tweeter (depending on source and music). I had the switches set to -3dB for the UMR and Tweet. I began experiencing a sound like an 'echo'. I put all the switches to 0 dB and that issue went away. I suspected a bad resistor(s) but now I'm wondering if perhaps one of the switches was 'hung-up' in a dual position thus causing 2 signals. I also recapped some 91s with Dayton PP film and was thinking of adding a series resistor to the tweeter cap. I've been listening to them lately and I'm kind of liking it as is. The bass ain't what the 90s have - but a very nice 3-way.

[DavidR]

On 11/27/2016 at 0:14 PM, ar_pro said:

I am currently using the Erse resistors in my AR-3a/AR-91 hybrid; their value is right on the money, which is more than could be said for the original AR-91 resistors.

AR-3a/AR-91 hybrid ?

Did you need a 2.5 ohm? I didn't see one in their line-up. Did you use a 2.7 or parallel (2) 5 ohm 12W ?

[ar_pro]

23 minutes ago, DavidR said:

AR-3a/AR-91 hybrid ?

Did you need a 2.5 ohm? I didn't see one in their line-up. Did you use a 2.7 or parallel (2) 5 ohm 12W ?

Well, it's sort of a hybrid.

Technically, it's an AR-91 - or maybe more technically, an AR-58, since it won't have level controls - inside of an AR-3a cabinet, with AR-91 crossover and drivers. The only resistors needed were the 1 and 2 ohm types used with the AR-91; the Erse 25 watt resistors measured spot-on.

You are exactly right about the Erse NPE's, David. I had tried their NPE caps in the woofer circuit, and found the result a bit "woolly"; both the 100 and the 40mF caps used in the AR-91 crossover measured about 10% high, while their Pulse X and PEx caps were right on value. I bit the bullet, and ordered Pulse X capacitors for the woofer, and got a much more pleasing result.

[DavidR]

Very interesting project. I hope you share your 'journey' with all of us.

[DavidR]

I recently heard of Meniscus Audio and was cruising the site and noticed they have MoX (metal oxide film) resistors by Lynk. They look just like the Mundorf MoX resistors but are half the price ($1.27 each) and they offer many more sizes than Mundorf on Madisound.

Does anyone know how these would stack up against the cement wire wound type and the pricey Vishay-Mills or Mills in a speaker system?

At 10 watts you'd have to parallel them and get 20 watts power handling.

[AR surround]

In my post of February 8, 2017 in this other thread:

http://www.classicspeakerpages.net/IP.Board/index.php?/topic/9293-ar-9-ar-90-crossover-re-capping-adventure/&

I state that I have found the proper balance among the drivers in my polypropylene film recapped AR90/AR9 surround sound system using the following settings on the stock attenuation switches:

Tweeter:-3dB; UMR -6dB; LMR 0dB

I have considered replacing a number of the original sand cast resistors with 1% low inductance units.  (Let's forget about using 0.3 ohm resistors to get back the ESR of the original NPEs as that little amount of attenuation isn't going to get me where I want to be.)  Given that it is a bit of a chore to get at those attenuation resistors, I have the following questions:

1. General: Has anyone achieved noticeable improvements in sound quality by removing the attenuation switches and associated resistors from the signal path altogether?

2. Assuming the answer to question 1 is "Yes," and that I have preferred the -3dB setting on the tweeter both prior to and after recapping, should I install a 2 ohm resistor before the 4uF cap and bypass the attenuation circuit altogether?  The tweeter crossover caps are Dayton 1% PMPC and don't seem to need any "burn-in."

3. The UMR is a bit more complicated, but again assuming that the answer to question 1 is "Yes":

3a) Should I install a 4 ohm resistor prior to the Solen 24uF cap and bypass the attenuation circuit?  My concern here is that Solen capacitor "burn-in" is not a myth and I will likely prefer less attenuation in the future.

3b) Perhaps I should install a 2.5 ohm resistor in the main circuit prior to the 24uF Solen and keep the attenuation switches?   That would give me the full 4 ohms / 6dB of attenuation that I currently use by setting the UMR switch in what is labeled the -3dB position.   But it also allows me some upside if I want less attenuation.  Downside is that it still keeps the aging switch network and a resistor in the signal path.

Thanks again for your valued opinions and advice.