Capacitors...

[Earl K]

Hi

MM = metallized mylar
MPP = metallized polypropylene
MPS = metallized polystyrene
PP&F = polypropylene and foil
PS&F = polystyrene and foil

I need some education here. Does " film and foil" mean separate alternating layers of foil between the plastic layers - within the construction of the cap - as opposed to many metal-backed plastic layers, for "metallized" ?
Or is it something else ?

regards <> Earl K

[Earl K]

Hi Ian

Thanks for sharing those thoughts about your capacitor induced insomnia.

For days I had sleepless nights at the rasping sound from the horns. I could just hear Jon saying..told ya so those 2307's........Seas shell sound...well this was way worse.

Imagine what will happen to your sleep patterns if you ever move to a bigger diaphragmed driver ( such as the 2430 ) and then remove those plastic lenses . I'm afraid even the Hovlands may not save your ears .

- Of course, that's just idle speculation. Plus JBL doesn't seem very interested in selling these drivers . ( so in that spirit, I think the Altec 288-8K is a much better choice in the 3" category for a few reasons - though its' massive magnet size can be off-putting for the DIYer )
- just me tweeking JBLs' nose ,

regards <> Earl K

[Ian Mackenzie]

Yeah I hear ya but.......,

But I would like to hear the 4344mk11, even if my ears have to be pegged back like a Rabbit..Muhahhahaha.

But seriously, my earlier 4343 diy of 20 years ago sounded a bit brutal in some ways....the diy crossovers used metallised polycarbonate capacitors.

The glare often associated with rigid metal radiator can also be directly attributed to solid state class A/B amps.

Sorry I'm on my soap box but hear me out thanks......

I started to appreciate this when I moved to pure class A about 3 years ago. The issue is the switching behaviour of the output stage of conventional A/b amps, no matter how well arranged the recurring odd order harmonics (of a series of aberrations) are particularly notice able to the ear.

The use of feedback to tame the above problem is also a sin the Pope would never Condon for it sucks the lifelike tonal character out of sound and contributes to the harshness by recirculating the offending damaged siginal .

There is also an apparent loss of very fine details or low level resolution.

Sounds like HiFi twackery I know, but I tried a dozen amps in a hifi club shoot out and they all produced some degree of ear bleeding compared the current reference diy X Aleph 100+100)

Some of you may recall the Stasis Theshold amps noted for their smooth clean sound back in the 70's & 80's. Well those designs are claimed to be a bit analytical compare the the current crop of X Series and earlier Aleph offering from the house of Passlabs.

A recent review in diyaudio.com.

The X Aleph had no problems at all man handling the Greg Timbers (JBL legend) 4345 JBL design with ease. The bass was a real surprise, it rendered space and timbre with accute accuracy and power. The vocals appear to be more visable to the eye and ear and all is revealed in the mix. The Micro dynamic resolve and instrumental clarity is spine tingling via the super 10 inch 2122 mid cone.

I find this quite exciting, at last an amp (a system) that allows you to view the portrait as is....... face on the canvas.

I'll get of my sop box now ...Muhahahahahahaha.

At least the black Art and Romance still lives with our favourite past time. Here his a recent picture of Pass taken at the CES.

Ian

[Earl K]

Hi Ian

But I would like to hear the 4344mk11, even if my ears have to be pegged back like a Rabbit..Muhahhahaha.

Well, I doubt if it would ever to come to that . I'm pretty confident that JBLs' newest implementation of LCR resonance control circuits ( K2 style ) would have taken out most of the objectionable things that we are now talking about. Ever since they've allowed the public ( in the orient ) to listen to unfettered horns they have really paid a lot of attention to smoothing out the response anomalies .
I wonder whether the S9800 was smooth enough to be able to discern a lot of front to back depth. Jons' testamonials have usually included some mention of ear plugs or the need for ear plugs - some ones idea of listenable ( at JBL ) must detour dratiscally from mine. Jon ?

Anyhow, I really do appreciate the soap-box chat.

And I realize that I've opened a veritable Pandoras' box in looking for driver fixes from simple filtering capacitors. Really, I'm just slogging my way forward knowing that eventually I'll likely have to learn how to ( efficiently ) implement those forementioned LCR control circuits. Until then - I'm exploring all the other remedies to a compression drivers obnoxious qualities .

Unfortunately ( for my ears ) most of my sonic explorations are driven/handcuffed by the need to make them appropriate or applicable in the SR world that I service. That's always present in my brain. Hence, some of my odd choice for eletronics - Behringer ? - really now !.

Secretly, I'm a closet "coner"- no jokes Bo. A few years back I had on the drawing boards my version of those large SOTA monitors ( made here in Ottawa - you posted a picture of one in the 43xx thread ). When Dynaudio stopped shipping components ( the same month I went to order parts ) - the project was shelved ( likely permanently ) .

I'll eventually move to class A amps - if only to give my ears a rest from all the "timing errors" that I'm presently immersed in.

FWIW, paper & oil caps sound pretty odd when "double" bypassed .

The X Aleph had no problems at all man handling the Greg Timbers (JBL legend) 4345 JBL design with ease. The bass was a real surprise, it rendered space and timbre with accute accuracy and power. The vocals appear to be more visable to the eye and ear and all is revealed in the mix. The Micro dynamic resolve and instrumental clarity is spine tingling via the super 10 inch 2122 mid cone.

I find this quite exciting, at last an amp (a system) that allows you to view the portrait as is....... face on the canvas.

A very nice quotation - thanks !

regards <> Earl K

[luxmanlover]

In an effert to smooth out my very hot sounding Eminence CD's I tried using DC rated PIO caps,( Dr. Bruce Edgar claims they are the ticket for taking the edge off HE systems) of a salvage sort.
In the case of the Eminence piece, success was elusive, this was the fault of the driver and not the caps.
Just for reference I'll post a pic of this monsters for anyone who hasn't seen one.
The caps on the left are 8uf while the middle ones are a 22uF Solen and the stack of PIO it took to make up an equivelent 22uF on a test XO. The Caps on the right are .33uF Solen and Jensens. The monster in the back is a 16 uf!!!! 4500 DC rated unit. I'd be interested in seeing a pic of those Mallorys you liked Earl, if possible.
Kelly

[Earl K]

Hi Kelly

Here's the pic that you asked for.

- there's a large 50 uf PIO cap in the back,
- then there's the 20 uf PIO cap somewhat centered - these are both G.E. types .
- the 17.5 uf Mallory metallized PolyPropylene is the smaller round can . It has a Mallory polystyrene cap with another small value PP bypass cap. The "spoiler" is a Seimens ( Europeon ) Metallized Polyester ( Mylar ) cap. It's 4.7 uf. It's part number is 21FD37175. But Mallory caps is now owned by Vishay Sprague and I haven't been able to determine what has happened to this line of caps.
- The 22 uf Black cap with the red epoxy ends has been my main DC blocking cap. It's of unknown lineage and must be a Mylar type due to it's sonic signature. I make up 3 packs of these things ( as you can see ) and can easily create large uf numbers.

- The blue cap ( epoxy dipped type ) is another sort of unknown , though it's made by SunCap. - 1.5 uf . As a single or in pairs and parralleled with an appropriate resistor ( & more bypass caps ) - it constitutes my only HF response contouring . I'm beginning to believe this SunCap is a metallized polypropylene - only because it's more focussed than a comparable Philips type Mylar cap in the same job funtion. Just guessing though. Might be sharper sounding Mylar for all I know .

Oh yes , I forgot, the obvious Solens coupled with some green RC ??? Mylar types. ( Not enough Mylar in that mix )

The inclusion of the B&C horns / DE16 driver are for todays amusement / the
2404 tweeter is for a size reference .

regards <. Earl K

[Robh3606]

Well I will finally be able to add my 2 cents on this. Just got the time to build my charge coupled 4435 compensation circuit. I am using Solens for the first run cause thats what I have on hand and on the second try it will be Mylar and Metalized Polpropylene no Solens to see what the deal is. Both networks will be charge coupled and the Solens won't be bypassed based on Giskards experience with them. I have an unbypassed Solen in there now. Never tried to do this so I am curious to see what if anything I will hear. Not saying there is no difference just not sure how subtle the effect will be and not sure what to expect. Hope the parts will be here before the hurricane

Rob

[luxmanlover]

I want to try a passive 2 way crossover between my 2123's and 2421/2470's as a possible alternative to tri-amping so I'm glad to see that I'm not the only one that has messed around with something that looks more like a bomb than a XO. I found a lot of PIO caps in surplus stores so it would be fun to be able to put them to use. Were any of the caps you used DC rated. Supposedly the dc rated ones are superior but if you had favorable results from regular AC caps it would make sourcing much easier. The place I found most of this stuff had literally tons of AC Mallory caps.
Kelly

[Alex Lancaster]

Aren´t Your caps the ones that got Lybians in trouble and a few people dead?

"Dayton" has been mentioned, are those out of Grainger?, We are getting Heavyyy Duty.

Alex.

[luxmanlover]

Well....I guess if one of those old babies blew there might be enough PCB laced oil flying around to cause a little carnage. Good thing they are rated at 1000VDC.
Kelly

Actually I had a motor start cap go off right beside my melon, it wasn't pretty.

[Alex Lancaster]

Luxman:

Years ago, I was in this business of making, selling, installing etc. Disco Stuff, We hired Engineering students, so, this really nice, hardworking, etc. nerd had the habit of hanging his head over Your shoulder, buzzard like, when You were really involved doing whatever, My partner who had a really wicked sense of humour, connected a polarized paper cap backwards, it exploded, and our nerd had His face covered with paper curls, too bad We did not have a camera, Party time!!.

Alex.

[Ian Mackenzie]

Earl,

Thanks for the feedback.

Some more Musings on Caps....... 1/2 soap box...Muhahahahah

The problem here is we all have our own rooms, systems and speaker configurations all with differing xovers and resulting sounds so its hard to draw conclusions with our postings.

My own observations recently however suggest " quality film foil capacitors are vital for hi end performance". If you can't do that expect compromise in the subjective appraisal.

But I deeply respect experiementation to obtain a great result using cascaded caps, its fun and you can blend the sound to your own taste and system.

IMHO if you can hear the differences, the driver/horn is accurate and the fault lies elsewhere as I found recently. Soft domes and curved doped cones far more forgiving.

Up stream.......................... a short story......stop snoring.

Previously I tried quite a few coupling caps in my Phase Linear 3300 Series 11 preamp. The metallised Solen fast were the worst. I eventually used better Burr Brown chips(OPA 2132) with low offsets and removed the caps all together.

Finally, after long and painfull saga I realise the harshness of so called horns ( read JBLs) was in part due to the chip ics. I now run discrete class A stages in all preamplification.

(pro equipment uses chips (or are discrete) but they are biased into class A and designed for 24 db of headroom)

It would appear hard domes/diaphragms cop a lot of flack coz they are accurate being pure piston range drivers and reveal the crap in the domestic audio chain.

We are hardly domestic so woopy do.

Recently experiments here at macka's Juicy Mansons suggest a clean crisp source is vital and noted the 2307/08 & slot discriminates markedly on variations in quality of the sources as they direct more DIRECT output to the listener than the room and hence portray more accuracy over a reverberant sound field of the 2344. (IMHO)

Why does the TV broadcast signal sound tizzy and crackly when 24/96 digital is pure bliss. One can only assume SACD is even better.

I mean hello it aint the horns Man.

Regards the 4344mk11, No wonder Japan is one of the last live Hi End audio markets where horns are highly sort after along with exotic SE amps.


Ian


My favourite JBL pic (outside Bo's living room) I'm sure this owner was pleased with these 4344Mk11's , although their sale remains a mystery.

[Earl K]

Hi Kelly

Well, ( tools of destruction? ) I can't help what my caps look like .

Were any of the caps you used DC rated. Supposedly the dc rated ones are superior but if you had favorable results from regular AC caps it would make sourcing much easier.

No , all my "motor run" caps are 50/60hz rated. I'll scout around and see if I can find any DC rated types.

Mallory made/makes ( according to an older catalog a couple of lines of AC motor run caps. They also made a couple of models of AC Motor Start caps - that I've heard one should avoid . Send me a PM if you need/want more info on Mallory model numbers for your shopping expeditions.

regards < Earl K

[Earl K]

Hi Ian

IMHO if you can hear the differences, the driver/horn is accurate and the fault lies elsewhere as I found recently. Soft domes and curved doped cones far more forgiving.

Up stream.......................... a short story......stop snoring.

Well, no snoring from me .

Here's a system testamonial ;

What I've cobbled together is certainly accurate enough to display the sonic signatures of different CD players / different electronic crossovers / and different amps - and now I find, different cap types. Any change in those areas is immediately apparent. When my system isn't in one of its' too often "deconstructed" modes, it offers near holographic imaging. It even somehow manages to extract depth from 1930s mono recordings of big bands. How, I have no idea.
When it's fully tweaked I can walk around the room and the imaging is sooo stable and real - that I appear to be walking around in a theatre balcony . ( The height thing is a consequence of my the systems vertical restrictions) But even within just a foot and centered between the speakers this sonic illusion is there - if not stronger - as I said - its' like peering down onto a stage area within a 1200 seat theatre. As I mentioned in a different thread - all this imaging is painted "behind" the speakers .

So why change it ?

I'm tinkering with my caps because they occasionaly get "rattled" from the huge dynamic swings presented by soprano voices - specifically those found in Verdis' ( opera ) Nabucco - Deutsche Grammaphon . Usually because of my line of work , I'd ask / demand the bloody sopranos rethink their stage blockings to give the microphones a break - but obviously "they've left the building" long ago . Too bad that, because they should have been told to take "three steps back" Divas !!!! Ha ! No problem from the Tenors .


Now because I'm a curious sort, I own a lot of the JBL/Altec componentry of the sort that we talk about - that's what drives some of my deconstruction .
When someone mentions something - I go to my Noahs ark of stuff and do my best to implement the topic. ( So much talk about the 2122 almost has me buying some. )

My thoughts on the 2344 horn and the 2307/2311/2308 combos are that generally they provide so much inherent mechanically induced filtering that a goodly portion of electrical/electronic filtering options/needs are negated. ie; they've had so much of the stuffing knocked out of them (by mechanical design ) that they'll blend/voice with just about anything. That's an overstated criticism/observation - and articulated to just make a point. Filtering can be mechanical or electrical. Right now I choose electrical .
My electrical filtering quest is to essentially to get back there from here but all the while preserving what those horns can't deliver.

regards <. Earl K

[Earl K]

Well it's certainly taken me long enough to get to this point.

First off; Wow !!!
Beautiful

The class A biasing of the capacitors is really the best thing I've heard in a long, long time .

I just have one side up and running. Additionally , the RC HF-boost/ midrange-suppression circuit is still running class AB.

One thing I like about this setup is the ease at which one can hear the different sonic signatures given off by the different dielectrics.

I more or less just jumped right in without following the planned path. Truthfully , ADD was setting in with the blending of cap types in the AB bias - and it was either jump ship now or leave town .

This bias setup is by far the easiest way to mix caps together to suit ones taste.

Of my caps tested;

(i) With regular type x Mylars ( 66 uf per side bypassed with polystyrene ) the signature from this cap appears as; "tightly wound" - lacking the ability to breathe much at all, but still retaining a nice midrange focus - so, lacking some warmth ( and sparkle ).
(ii) PIO - even just 20uf substituted into one side adds that ability of the lower mids to breathe and as a consequence seems to add a lot of warmth to the mylars ( a little sparkle ).
(iii) Polypropylene in oil - 17.5uf substituted in kept some of the oils warmth while enhancing the presence area - I guess because of the MPP portion of the cap.
(iv) Solens/Mylar mix - 29uf total (19uf of Solen content ) - substituted into one side opposite the 66uf three pack of Mylars . Well the sonic balance tipped immediately . Man those are "HOT" capacitors . Way too much presence. They were yanked in tens of seconds. I'll revisit them with straight paper in oil to see if there's any hope .


As I said , I'm just doing this in mono right now. The other side is a straight wire ( no DC blocking caps ) - with a RC network for HF enhancement. So I don't know what's happening for the total - in regards to front to back depth. There's such a speed difference between the 2 sides that right now that they appear as almost 2 separate sources.

- You'll notice I've paid no attention to maintaining symmetrical/balanced loads fo the cap values across the 2 sides. JBL runs this way ( unbalanced ) in a few places in their K2 networks. I suppose they felt the concept is inherently self balancing in an AC network ( even across inductors ).

- I'll come back to this later.

regards <> Earl K ( back to listening )