The No.1 Website for Pro Audio
Cap Upgrade - Main Speaker Crossovers
Old 18th June 2015
  #1
Gear Addict
 
🎧 5 years
Cap Upgrade - Main Speaker Crossovers

Hi People,

Just wanting to check with knowledgeble folk out there with regard to speaker crossover caps, I'm going to replace the stock metalized films with something better as theyre run of the mill... I wouldnt usually bother with something like this but I have heard crossovers are where you definately hear improvements in cap changes.. since there's only three in there I want some quality parts.. There's one question regarding electrolytics.. I discovered there are a single axial AC bipolar electrolytic in each,,.. yuck!! I wanted to check if using a very high quality radial part having to have long leads would be inferior to a so so axial parts with appropriate short leads..?? Only because i have found some black gates in the right size and theyre radial , Ive read great things about these caps but would the long lead length negate the quality upgrade? If I can find a film part in the same capacitance would that be even better than a blackgate?

p.s calling Jim Williams... Hi mate, can you tell me if the REL PPFM Audiocaps are decent? I was thinking of going with these despite the price... then again others go for way more but thats kind of my budget. I realize they all have a bit of a sound, but really would like the most invisible.. Any tips much appreciated!

Cheers.
Old 18th June 2015
  #2
Lives for gear
 
Silent Sound's Avatar
 
5 Reviews written
🎧 10 years
What kind of speakers are these?

Personally, I wouldn't worry about long leads. You're gonna have a long cable running from the amp to the speakers anyway. I don't see how one or two more inches of wire would effect much. Radial caps would be fine, though probably more difficult to install.

As for upgrading crossovers, you have to be careful. I've upgraded crossovers on speakers before and ran into the problem where the new crossover doesn't sound near as good, even if it performs better. For instance, you could find out that switching electrolytic caps for a film one lowers the ESR and increases the volume of the tweeter, ruining the balance of the speakers. Also, I've had the issue where you switch to better caps, only to realize that increasing the high frequency clarity was a bad idea because the tweeters don't produce those highs as well. It can make a warm speaker sound strident.
Old 18th June 2015
  #3
Lives for gear
 
JohnRoberts's Avatar
 
🎧 10 years
+1 be careful what you improve... The ESR of the old cap may be factored into the Xover design equations.

If you want to make it really better, bi-amp or tri-amp with electronic crossover, but that is not trivial to design.

Safest approach is to use similar cap to original equipment, just newer.

Maybe ask the manufacturer for advice.

JR
Old 18th June 2015
  #4
Gear Addict
 
🎧 5 years
Thanks guys, yeh this sounds wise.. I actually re-capped a tape machine once with ultra low ESR caps in the signal path., and the audio came back sounding strange..especially in the high-end.. kinda slightly 'scratchy'. Im guessing something there was ringing or something but whatever it didnt sound good.. perhaps this is what your talking about, cheers!
I may just go for it and be ready to put things back, the parts in there are cheap, and could get lucky (this sounds like a terrible approach but im no tech).. just want to get monitor chain clean as i can.

Can I ask John.. What are your thoughts on 'bypassing' these caps in signal path situations? I know there are some firm believers, and some not eg Steve Hogan.. that they can cause instability unless you know exactly what your doing again causing ringing/oscillation up high... i find it interesting but perhaps the issue is knowing exactly what to do and where.
Old 18th June 2015 | Show parent
  #5
Lives for gear
 
🎧 5 years
Quote:
Originally Posted by JohnRoberts ➑️
+1 be careful what you improve... The ESR of the old cap may be factored into the Xover design equations.
.........................................
JR
This is important and so often overlooked in crossover upgrades.
There is more to a crossover than just reading the value off the old part or reading the nominal value in the schematic.
Old 18th June 2015 | Show parent
  #6
Lives for gear
 
JohnRoberts's Avatar
 
🎧 10 years
Quote:
Originally Posted by PartHunter ➑️

Can I ask John.. What are your thoughts on 'bypassing' these caps in signal path situations? I know there are some firm believers, and some not eg Steve Hogan.. that they can cause instability unless you know exactly what your doing again causing ringing/oscillation up high... i find it interesting but perhaps the issue is knowing exactly what to do and where.
I have written about this more than a few times since the '80s but at the same time caps have changed quite a bit since then too (mostly for the better but not always).

As a circuit designer myself I do not encourage people second guessing design choices without complete information.

While my personal bench research was decades ago, using parts from back then, I found that I needed to use at least 10% of the total capacitance in my parallel film caps for the combination to act like a big film cap.

Much easier when doing a scratch design to manage where the caps can be important, and make those caps good... For example I've used millions of electrolytic caps in console signal paths, but i generally tuned the poles to be well below audible response so a single HPF tuned higher using a film cap could dominate the path response and sonics.

This is simplistic but in general trust a good design engineer's choices and replace his cap choices with modern equivalents. If second guessing his design try to REALLY understand what you are doing.

JR
Old 19th June 2015 | Show parent
  #7
Lives for gear
 
lame pseudonym's Avatar
 
🎧 15 years
Quote:
Originally Posted by JohnRoberts ➑️
While my personal bench research was decades ago, using parts from back then, I found that I needed to use at least 10% of the total capacitance in my parallel film caps for the combination to act like a big film cap.
Long ago I recapped a Tascam 388 with Muse capacitors (considered hotsy-totsy then), and found that the leading edges of the music were so obscured that I couldn't play new tracks in sync. After a little experimentation I landed on the conclusion that I had to calculate film bypass values to reach down at least to 250Hz if they were the only capacitors in the position. It was a chore to add all of those but let me tell you, when I was done it sounded nice.
Old 19th June 2015
  #8
Gear Guru
Most speaker crossover parts are the cheapest parts they could buy as you never see that stuff. Sand resistors, steel core inductors, electrolytic or mylar film caps, etc.

I have more $ invested in my speaker crossovers than I spent on the speakers new. Copper foil inductors are selected to match DC resistance and inductance. I use exotic film caps from MIT, InfiniCap and Mundorf. Resistors are Mills MRB wirewound.

Internal wiring is Kimber AGSS pure silver/teflon. Yes, it was worth it. You would be lucky to find such parts in $5000+ speaker systems.
Old 19th June 2015 | Show parent
  #9
Gear Addict
 
🎧 5 years
Quote:
Originally Posted by JohnRoberts ➑️
I have written about this more than a few times since the '80s but at the same time caps have changed quite a bit since then too (mostly for the better but not always).

As a circuit designer myself I do not encourage people second guessing design choices without complete information.

While my personal bench research was decades ago, using parts from back then, I found that I needed to use at least 10% of the total capacitance in my parallel film caps for the combination to act like a big film cap.

Much easier when doing a scratch design to manage where the caps can be important, and make those caps good... For example I've used millions of electrolytic caps in console signal paths, but i generally tuned the poles to be well below audible response so a single HPF tuned higher using a film cap could dominate the path response and sonics.

This is simplistic but in general trust a good design engineer's choices and replace his cap choices with modern equivalents. If second guessing his design try to REALLY understand what you are doing.

JR
Right, thats interesting. So am I right in saying, your effectively using the electrolytics to do theyre job but pushing their bandwidth lower than 20Hz.. and letting the film part pass the audio with its purer tone in the human hearing range?? Even if thats not the case what you said sounds clever! Well Im not going to even try it as I dont know where to start with calculations, no test equipment etc.. so yes may just replace one off parts with high quality equivalents. May make a lil difference for cleaner monitoring
Old 20th June 2015 | Show parent
  #10
Lives for gear
 
JohnRoberts's Avatar
 
🎧 10 years
Quote:
Originally Posted by PartHunter ➑️
Right, thats interesting. So am I right in saying, your effectively using the electrolytics to do theyre job but pushing their bandwidth lower than 20Hz.. and letting the film part pass the audio with its purer tone in the human hearing range?? Even if thats not the case what you said sounds clever! Well Im not going to even try it as I dont know where to start with calculations, no test equipment etc.. so yes may just replace one off parts with high quality equivalents. May make a lil difference for cleaner monitoring
I wrote a piece about this back in the '80s for my audio mythology column, while a simplification most of the non-ideal behaviors in electrolytic caps are related to frequencies when they express significant terminal voltage (at tuning and below). If you set the pole frequencies for electrolytic blocking caps to be very very low frequencies, they never develop any terminal voltage to cause trouble (that was easy).

This is one of those raise the bridge or lower the water design trade-offs. I have used many electrolytic caps in audio paths, but never to form a pole at an audible frequency. i use film caps for audible poles.

JR
Old 20th June 2015
  #11
Gear Guru
Back in the 1970's it was common for designers to use a roll-off blocking cap set to 20 hz or slightly below that. What they missed was the accumulation of poles, one for each capacitor used. The group delay was significant as all those poles do add up.

Walt Jung suggested a roll-off frequency at about 1 hz to avoid that build up. Phase shift kicks in a decade above that roll-off point so any cap operating at 20 hz creates a phase shift curve beginning at 200 hz. Roll-off at 2 hz and the phase shift remains below 20 hz, out of the audio band.
Old 21st June 2015 | Show parent
  #12
Lives for gear
 
JohnRoberts's Avatar
 
🎧 10 years
Quote:
Originally Posted by Jim Williams ➑️
Back in the 1970's it was common for designers to use a roll-off blocking cap set to 20 hz or slightly below that. What they missed was the accumulation of poles, one for each capacitor used. The group delay was significant as all those poles do add up.
They.....
Quote:
Walt Jung suggested a roll-off frequency at about 1 hz to avoid that build up.
I don't recall targeting a specific frequency for the very low electrolytic poles, just << audio. You probably don't want to know how many HPF poles a signal encounters from input to mix down in a console. But customers won't tolerate scratchy pots and switches that click, so instead they get many HP poles in series. This can get difficult to keep << everywhere. Circuit nodes like the cap in series with a mic preamp gain pot could have a 10 ohm or less end limit resistance. That involves a pretty large capacitance to tune <<..

One dirty circuit trick used by at least one competing product I dissected used an undersized cap in the gain pot leg while specifying a full 20Hz bandwidth... It may have been 20Hz at low gain but at max gain the HPF pole was well up into the audio band, and their sheeple customers believed the hype that their preamp was quieter than the others with honest frequency response.
Quote:
Phase shift kicks in a decade above that roll-off point so any cap operating at 20 hz creates a phase shift curve beginning at 200 hz. Roll-off at 2 hz and the phase shift remains below 20 hz, out of the audio band.
Those are just arbitrary rule of thumb but yes it is all cumulative and one pole sections exhibit gradual continuous phase and amplitude slopes. If the half power -3dB poles are set for 20Hz and 20kHz the amplitude will be down -1dB at 40Hz and 10kHz, -1/10 dB at 200 hz and 2kHz... Conservative design sets the HF half power tuning for 200kHz or higher. I was comfortable putting real LF poles at 20Hz since too much low bass content wouldn't make it though magnetic tape paths, and could damage speakers. Now using modern digital paths an argument could be made for extended LF response, while the HF response will be dominated by conversions to and from digital media.

+1 it all adds and it all matters.

JR
Old 21st June 2015
  #13
Gear Guru
In the 1970's console designers were all over the map as to how much low end extrension was needed. You would find a Neotek series II with 10 uf caps hanging on opamp outputs, a real problem if interfacing with 600 ohm transformer coupled equipment. Then again, lower cost analog consoles like Tangent used 220 uf caps almost everywhere and those never had the thin or sloppy low end the Neoteks had.

These days, I select a 2 hz roll-off to get flat phase lines on the AP analyzer, when I use a blocking cap. Mostly I've managed to avoid using DC blocking capacitors in most of my recording gear and that has improved audio quality enormously. The best sounding capacitor I've ever heard was a piece of wire.

Ray Kimber was doing some research on developing passive speaker crossovers using resistors and coils only, no capacitors. They sounded pretty good to me.
Old 22nd June 2015
  #14
Gear Addict
 
🎧 5 years
John and Jim very interesting, almost like youre steering the cumulative nasties (which there will be) away from the audio band.. I was reading a post from John Curl of Parasound/Wilson audio, he was saying in his amps if there isnt budget constraints he will use interstage servo's wherever he can to avoid this stuff.. what are thoughts on servo's instead of electrolytics... do they have any negatives or trade offs also???

Thanks
Old 22nd June 2015
  #15
Lives for gear
 
🎧 15 years
Hi
Servos are just a 'sidestep' and can introduce more 'problems' than they solve.
If used in 'sensible' places they can be a handy tool but as they add another op amp and a small handful of passive components. They can certainly be a great help where ordinarily a large electrolytic may have been used, perhaps around a gain pot in a mic amp.
Matt S
Old 22nd June 2015 | Show parent
  #16
Lives for gear
 
JohnRoberts's Avatar
 
🎧 10 years
Quote:
Originally Posted by PartHunter ➑️
John and Jim very interesting, almost like youre steering the cumulative nasties (which there will be) away from the audio band.. I was reading a post from John Curl of Parasound/Wilson audio, he was saying in his amps if there isnt budget constraints he will use interstage servo's wherever he can to avoid this stuff.. what are thoughts on servo's instead of electrolytics... do they have any negatives or trade offs also???

Thanks
Funny I wrote about DC servo's in my 1980's magazine column too...

The short answer is servo's have their place, but IMO that place is not every place, due to cost, and path complexity. I also believe in KISS (keep it simple).

Servos only became practical with the rise of op amps that don't suck (mid/late 1970s). Combining a good op amp with high impedance resistors and a small high quality film cap we can block DC with far less non-ideal behavior.

I have used exactly one DC servo in several decades of circuit design, and that was for an over-engineered phono preamp that used a normally difficult to DC couple FET front end gain stage. Integrating a DC servo into the current source biasing the JFET front end I was able to mange the DC operating point of the front end nicely. I still included a film cap real one pole HPF after that.

So I have been content with using electrolytic caps in the audio path, as long as they are tuned low enough that normal audio does not generate any terminal voltage across them.

That said a good friend of mine who was chief engineer for a well respected analog console maker used a handful of servos in his console signal path (around the fader/channel mute stage for example). His consoles were very well regarded for their sound quality so when properly executed servos can sound good, or do no harm.

Finally all servos are not created equal. Topology around the servo matters and I prefer not exposing the input (or the output) of the servo to high edge rate signals. Especially if using op amps that are more optimized for DC performance than AC performance. So there is some art involved in designing servos too.

Servos are a little like gold plated audio jacks. All else equal they generally don't hurt, but do not guarantee that the rest of the design is optimal, only that the designer threw some extra money at it.

JR
Old 24th June 2015
  #17
Gear Addict
 
🎧 5 years
Hi John, awesome.. It all sounds like harder work but one of those money and complexity things, probably worth it given budget and design goals. I'm having my console re-done right now, the tech said to me the problem with many consoles like mine is lots of FET switching that all adds up dirt...perhaps these are good spots for servo;s like your friend does, or even just no switches. I think GML uses them in some of his pieces, that small batch HRT9100 line mixer looks amazing, the specs are pretty insane.. I'm sure hi res analog never comes cheap though.

To you, Jim or Matt... Can I ask, I've been looking at these monitor crossovers and theres only two caps. 1 x 20uf non polar electrolytic for the main driver, and 1 x 3uf polyprop film for the tweeter. I've looked online at their specs and none of them post ESR, all I can find is a spec Dissipation Factor, in Ohms. I'm contacting the manufacturer just to make sure there were not critical factors like this designed in to the voicing of the speakers.. Is this likely the function that they could have used, if so can I just match that spec or is there more to it? , are there any other parameters I need to ask them about??

Also regarding the electrolytic, I can find a few film parts in 20uf but would like to avoid metalized polyprop and get some tin foil parts for lower DA.. No parts that big, Is it at all possible to 'piggyback' 2 x 10uf parts to reach the value?? If so any tradeoffs to a technique like this?

Again many thanks and kind regards

Last edited by PartHunter; 24th June 2015 at 03:29 PM..
Old 4 weeks ago
  #18
Here for the gear
 
pablometal's Avatar
Hello everyone, don't wanna open a new thread since we already have this one. I wanna ask a quick question, is it worth to upgrade the caps of my T7V's? I have tools for doing it and basic knowledge required for the task, bought these speakers around 3 months ago. I like how they sound but i think they could sound better (cleaner).

What capacitor brand would you recommend?
Old 4 weeks ago
  #19
Here for the gear
 
pablometal's Avatar
upppppppppppp
πŸ“ Reply
Topic:
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump