leok Posted December 22, 2003 Share Posted December 22, 2003 I had the following in the odds & mods section for a couple of days with no response at all. The intent here is not to change anyone's mind, but to propose an avenue for performance improvement that may resolve a few mysteries regarding cap specs vs performance. Thinking odds & mods types might be more interested in the technical .. since cap conditioning is a big pain .. I thought I'd try there. Maybe interest in improved sound, more a focus of this forum, will be cause for more interest. For my part, no interest at all is information as valuable as overwhelming interest. I believe that the electronics industry's habit of distributing products that are not, and may never be properly conditioned, is interesting behavior. What do the designers think about the fact that most customers may never hear their products as they intended? My RF-7s now sound fabulous, but they just were never going to get that way without some agressive corrective action. I suspect there are many Klipsch speakers out there in a similar state. But, the requirement that people either pay huge sums of $ for exotic caps, or perform a conditioning operation that only a skilled technician should attempt is too much to expect, and I think that goes a long way in explaining the indifference to this proposal. People are reasonably thinking "You've got to be kidding." ------------------ While the best-crossover-capacitor controversy goes on I have found a well known characteristic of plastic (polypropylene, mylar, etc.) caps in particular, but of all caps to some degree, that I find swamps the actual differences between the performances of the various technologies. The characteristic is conductive defects in a new cap that have not been conditioned by exposure to high voltage. Shortly after purchasing my RF-7s I replaced what looked like mylar, stock crossover caps with polyprop. film/foil. The sound improved. It was smoother with less high frequency distortion. Recently I replaced those polyprop. film/foil caps with very big and expensive paper in oil caps. The improvement in sound .. the reduction in high frequency distortion was tremendous. It was more tremendous than made any sense to this engineer .. I felt there must be more to the story than some PIO magic. In a subsequent experiment I conditioned similar polyprop. film/foil caps (and a stock low pass n.p. electrolytic) that I had placed in the crossovers of my Chorus-II speakers. This conditioning involved application of about 35VDC, in both polarities, to all the caps in the crossover. The improvement in Chorus-II performance was on the same scale as the improvement that PIOs had made in the RF-7s. This time, no cap change, just cap conditioning. Crossover caps see very low voltages and, in Klipsch speakers, very low power. The voltage and power is simply not enough to effectively "break in" the caps. The defects that would normally be vaporized by the first application of working voltage are never vaporized and the caps continue to conduct currents via these defects. As a result, the un-conditioned crossover caps are actually a combination of capacitor and nonlinear resistor. I suggest the following procedure for the RF-7 speaker. * Secure a DC voltage source: 35V to 75V. * Insert a 3.3KOhm resistor in series with the voltage source. * Disconnect inputs and all jumpers from the 4 RF-7 input posts. * remove the horn (this involves disconnecting the wires to the driver) * remove the upper cone driver (disconnect the wires) (with the horn out, you can reach in and support the cone driver as you remove the screws. Be careful. It's very heavy and expensive). * disconnect one of the wires from the lower cone driver. -----at this point ALL drivers are disconnected !!! ----- Using the DC voltage source in series with the resistance, apply voltage to the connections as follows. Connect one polarity for 20 seconds or more, then reverse polarity and apply for 20 seconds or more. * across the +/- leads that were removed from the upper cone driver * across the +/- leads that were removed from the horn * across the "HF" terminals at the rear of the speaker. Short all leads and terminals together to remove residual cap charge. * re-connect the lower cone driver. * connect and mount the upper cone driver * connect and mount the horn * re-connect input jumpers and amp The speaker sound should be much smoother. Since I had already replaced the stock caps in my RF-7s I don't know how effective this will be with the stock RF-7 caps. If there is an RF-7 owner in the Boston area who would like to try this, I will perform the complete operation at no charge (one customer and one speaker pair only) in order to determine if the stock caps will improve in a similar fashion. Leo Quote Link to comment Share on other sites More sharing options...
Klipsch RF7 Posted December 22, 2003 Share Posted December 22, 2003 Hi Leo, Would you mind if asked you a favor, could you please take a pic of the crossover for the RF-7's and point out which Caps need to be replaced? I would greatly appreciate it. Thanks Quote Link to comment Share on other sites More sharing options...
leok Posted December 22, 2003 Author Share Posted December 22, 2003 GFH, The point of my discussion about conditioning is that a fair portion, if not all (although I question that) of the gains made by replacing the caps may be possible by simply conditioning the ones that are in there. Secondly, Unless you use very expensive, and large PIO caps, even very good polypropylene caps will probably arrive from the manufacturer unconditioned, and your gains will be minimal unless those caps are conditioned. In the end, the PIOs probably also benefit from conditioning. I conditioned all caps in my RF-7s last Fri, and there is definate improvement beyond simply adding the PIOs (which itself was a remarkable change. Taking a photo at this point requires removing a cone driver (because the crossover is now distributed over inside cabinet real estate). If you have an RF-7, the caps, in their original size and position, may be seen by removing the crossover panel mounted to the back of the cabinet. Leo Quote Link to comment Share on other sites More sharing options...
Klipsch RF7 Posted December 22, 2003 Share Posted December 22, 2003 Is that write up for conditiong the crossovers accurate enough for me to print out and hand to a TV/VCR repair shop to do the work? Will this procedure work with the Auricaps? Why did you go with the Jensen instead of the Auricaps? Which one would be worth my time? Thanks Quote Link to comment Share on other sites More sharing options...
leok Posted December 22, 2003 Author Share Posted December 22, 2003 I went from stock to Hovland "Musicap" polypropylene fil/foil. The reason I went on to the PIOs was I wasn't satisfied with the results. However, had I known about the conditioning, I may have been perfectly happy with stock or the Hovlands. A good technician will be able to follow the instructions I entered for the RF-7s. Leo Quote Link to comment Share on other sites More sharing options...
Klipsch RF7 Posted December 22, 2003 Share Posted December 22, 2003 Nice, so is the procedure the same for Auricaps is what im asking Remember you told me to get the 12uf and 5uf caps for my crossover by Auricaps. Thanks so much for your expertise and all this tweaking, i love to tweak stuff and this sounds perfect to get every last ounce of sound from my Sprechers. Veilen Danke, Quote Link to comment Share on other sites More sharing options...
leok Posted December 22, 2003 Author Share Posted December 22, 2003 The procedure applies to Auricaps. You might try the stock ones first, but if you're going to insert the Auricaps anyway, then why do something painful twice (except to provide us all with some very interesting information)? Leo Quote Link to comment Share on other sites More sharing options...
Tubinhard Posted December 22, 2003 Share Posted December 22, 2003 I am not a "good" technician, so I need some specifics. "Secure a DC voltage source: 35V to 75V." Please define a DC voltage souce that would generate those voltages. Short all leads and terminals together to remove residual cap charge. Since we have wires from the voltage source hooked up to both positive and negative through this process (including reversing the polarity) -- what "leads and terminals" are we shorting, and once identified, how are they shorted? With that out of the way... My understanding is that PIOs are "lossy", that is -- they do not hold a charge for very long. The cap is charged, and then immediately begins to "drain". So, poor dielectric absorbtion properties. In fact, they are not doing what they should be doing. Any chance at all that this procedure may be causing a slight breakdown of the dielectric, resulting in some "lossiness"? Quote Link to comment Share on other sites More sharing options...
NOSValves Posted December 22, 2003 Share Posted December 22, 2003 Dean, My understanding is that PIOs are "lossy", that is -- they do not hold a charge for very long. The cap is charged, and then immediately begins to "drain". So, poor dielectric absorbtion properties. In fact, they are not doing what they should be doing. Any chance at all that this procedure may be causing a slight breakdown of the dielectric, resulting in some "lossiness"? I think you may be reading just a little to much into what my testing was showing with my new cap analyzer. I wouldn't go as far as the post above suggested. In fact at this point I really can not make any clear assumption about what the different caps do and don't do I just know the Jenson caps I tested act different then the others I was putting thru the same testing procecure. Craig Quote Link to comment Share on other sites More sharing options...
Klipsch RF7 Posted December 22, 2003 Share Posted December 22, 2003 Ok, if these instructions are precise then ill save them and use them to give to a tech to do the procedure, i just hope there are not any grey areas or any bit of info left out. Quote Link to comment Share on other sites More sharing options...
Tubinhard Posted December 22, 2003 Share Posted December 22, 2003 Oh, I didn't get that information from you -- I found it out on the web somewhere. When you get time to play, it will be interesting to see what kind of conclusions you come to. Quote Link to comment Share on other sites More sharing options...
leok Posted December 22, 2003 Author Share Posted December 22, 2003 Dean, Any voltage from 25V to 75V will work. How to generate .. this is one of the problems. I'll send email. Short the two (+/-) horn leads together. Sort the two (+/-) cone leads together. Put a wire across the HF input terminals. Can't help you with the PIO question. If what you say is true, the losses just aren't significant in applications in which I've used them. All caps I've seen in Klipsch crossovers are rated at 100V or greater. If you've substituted any of your own you have to look. None of the replacement caps I've mentioned are rated under 100V. If the dielectric does break down then the cap was faulty to begin with. Quote Link to comment Share on other sites More sharing options...
Tubinhard Posted December 22, 2003 Share Posted December 22, 2003 Interesting thoughts about the inductors -- hmmm, wouldn't that suck. Quote Link to comment Share on other sites More sharing options...
stream Posted December 22, 2003 Share Posted December 22, 2003 I'd love to try this, but I'm quite sure that I would do something wrong -Jesse Quote Link to comment Share on other sites More sharing options...
Klipsch RF7 Posted December 22, 2003 Share Posted December 22, 2003 Craig, If i was to pull out the crossover network and then ship them to you, Could you solder in the Auricaps and do the conditioning on them for me? Thanks Quote Link to comment Share on other sites More sharing options...
Tubinhard Posted December 22, 2003 Share Posted December 22, 2003 Craig sends all of the crossover work to me. Don't worry, I use the best super glue, ah, I mean, solder there is. Considering the price of Auricaps these days, I would go with a good film and tin foil instead. I would also yank one resistor in each board and replace it with a Mills. Quote Link to comment Share on other sites More sharing options...
leok Posted December 22, 2003 Author Share Posted December 22, 2003 Assuming cap repair charges would be small, risk to inductors would be even smaller. Not allowing repair charge to flow through an inductor would further inhibit a novice from repairing what the supplier should have taken care of in the first place. I doubt there is enough energy involved in these procedures to leave any residual flux in an inductive device. Note also that each procedure applies reverse polarity which will further reduce the possibility of residual flux. If you believe I'm wrong please respond. Leo Quote Link to comment Share on other sites More sharing options...
ben. Posted December 22, 2003 Share Posted December 22, 2003 Very interesting stuff, Leo. I have nothing to add, but I thought I'd say that I'm paying attention. Quote Link to comment Share on other sites More sharing options...
3dzapper Posted December 22, 2003 Share Posted December 22, 2003 Here's a down and dirty way of getting DC. Build a board with two junction blocks, run two 18" wires terminated with insulated alligator clips to two opposing blocks where you will put the cap and a 1kohm resistor between two more then jump those two together. Hook the leads to the plate of a spare preamp (about 100vDC), disconnect the leads for a minute and connect them in reverse. do this a few times and you are done. Just be careful. Remember: one hand for the amp and one hand for the back pocket! Rick Quote Link to comment Share on other sites More sharing options...
leok Posted December 22, 2003 Author Share Posted December 22, 2003 Ryan, I think we see the same potential problem and solution. Leo Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.