That is a 1st order bandpass filter on the mid driver at some odd frequencies and they have reversed polatiry of it... not sure that would make any differnece, and hung the piezos of that too with an additional resistor inline. I'm not sure about the logic of any of this, the mid horn is more sensitive than the woofers and the piezos so it needs a pad not the piezos, and a 1st order filter is woefully inadequate on that compresion driver, it should have a 2nd order or even a 3rd order high pass filter, if you wanted to simplify a bit just lose the lowpass filter for the CD, just let that driver go up as far as it can.
Just to be clear, that schematic I drew is of the original crossover. Yeah, I couldn't find any other three-way designs that just slapped the woofer(s) across the jack and hung everything else parallel to that. I was trying to keep in mind that the original designers knew there would be piezos. They also apparently knew that the original one-piece horns (Heppner - similar to many found in Magnavox console stereos) had a higher impedance than everything else; I could never find a rating for them but their DC resistance was around 15.6 ohms.
Amazon is reporting that my shipment that includes the updated compression drivers appears to be lost, so I canceled the order, requested a refund, and placed the order again. So right now, my planned order of ops is:
Amazon is reporting that my shipment that includes the updated compression drivers appears to be lost, so I canceled the order, requested a refund, and placed the order again. So right now, my planned order of ops is:
- Receive and install updated compression drivers
- Put 10-ohm/10-watt resistor across the piezos
- Test spectrum with EQ flat to find best adjustment for L-pads on the mids
- Test phase between horns and 12"s at 500Hz to find best setback distance for horns
- Build mounts for horns per setback distance
- Reproduce previously-found L-pad setting with fixed 20-watt resistors and attach to mounts
- Test spectrum again (preferably far-field) and adjust graphic EQ accordingly
Do you now have the 12" driver wired in series for an 8ohm load to the Dayton crossover?
Piezo drivers have a very high impedance.. like 1000ohms, this is why it was claimed that they can be used without any crossover at all. That is true at lower power levels but in my experience they sound better and last longer in PA applications when a crossover is used. It is also recommended to put an 8ohm resistor across them to present a steady load to the amplifier and passive crossover.
Check out the thread linked below for some more info on piezos.
https://audiokarma.org/forums/index.php?threads/using-piezo-tweeters-wisely-a-how-to.79659/
Piezo drivers have a very high impedance.. like 1000ohms, this is why it was claimed that they can be used without any crossover at all. That is true at lower power levels but in my experience they sound better and last longer in PA applications when a crossover is used. It is also recommended to put an 8ohm resistor across them to present a steady load to the amplifier and passive crossover.
Check out the thread linked below for some more info on piezos.
https://audiokarma.org/forums/index.php?threads/using-piezo-tweeters-wisely-a-how-to.79659/
Yes, I was just giving my observations. That was certainly a minimalist approach that could work but not with the drivers the cabs were equipped with when you took posession, and even if it did work it probably wouldnt last long at the drive levels this type of speaker might see.
@conanski - And it just so happens that I received the cabs with one of the Heppner horns blown. Unable to source a replacement diaphragm/coil that would actually work so I gave up on the Heppner horns altogether. It was too bad, too, because the surviving Heppner horn sounded better and had a smoother response than either of the replacement one-piece 4"x10" horns I tried and ultimately discarded.
The crossover is designed for a nominal 8ohm load so that would be the optimum value to use to get the desired filter resposne... a 5khz high pass.
A pair of piezos in parallel still have an impedance of maybe 100ohms at 20khz and it's much higher than that down at 5khz, the amplifier doesn't know they exist, no current worth talking about is consumed the drivers merely respond to the voltage signal that appears across the resistor.
Last edited:
As I was alluding to above, it's likely the blown driver was a result of the previous owners DIY crossover attempt. One of the primary jobs of a crossover is driver protection... particularly the high pass portions for mid and hi frequency drivers, those filters have to be steep enough and at a high enough frequency to prevent over excursion, for that it's usually 2nd order at driver FS plus a bit at the very least, many will suggest 2x driver FS.
Two Piezo tweeters in parallel have an impedance @ 20Khz of roughly 35 ohms. ( +10ohms = roughly 8 ) I meant it is 40Khz where it gets really low.
By all means use a standard value of 8.2 ohms, but don't go lower 🙂
PS.
The 'electrical function' of a Piezo is actually that of a lossy capacitor.
Well, this was a smashing disappointment. I received the two updated compression driver units and put them to my "wig-wag" white noise test that pans signal between them and found that they also sound markedly different. Yes, I swapped the drivers on the wires to make sure the difference wasn't upstream; it's definitely the drivers. I'm fine with using the EQ to flatten out minor inconsistencies but this is too much to ask of it. I've contacted the manufacturer's support site to ask what if anything can be done but I think I'm looking at a return.
UPDATE: I put sinewave signals through the drivers at very low power levels and discovered that one of them has a very distinct "kazoo" distortion at about 1kHz and below. I found that I could affect the distortion by just squeezing the driver enclosure with my bare hands - maybe even to the point of making it disappear. This really does seem like a manufacturing problem. I've updated my support ticket with the manufacturer accordingly.
I found out many years ago that this distortion you describe is nearly always due to improper centering of the diaphragm into
the voice coil gap > either by poor manufacturing OR poor replacement of a diaphragm.
I also found out that Piano music is highly effective at exposing these distortions, rather than just a single frequency.
the voice coil gap > either by poor manufacturing OR poor replacement of a diaphragm.
I also found out that Piano music is highly effective at exposing these distortions, rather than just a single frequency.
Unfortuntely this is common with cheap compression drivers and even with cheap chinese replacement diaphrams in expensive drivers, compression drivers are precision devices with fractions of a milimeter clearance in some cases so the diaphrams must match the shape of the phase plug and be perfectly symetrical, there is no room for sorta-kinda good enough mass produced junk.
I think you’ll find the mid driver is a Selenium 250 and is only good down to 1k if you want power handling and low distortion. It’s probably the best of the cheap mid drivers out there but is not intended for critical listening, it’s for small PA work only. I wouldn’t bother replacing the diaphragm, just get original Selenium drivers.
@MusicLover - Funny you should mention that. The first thing I did after I did the wig-wag white noise test - and not deliberately, mind you - was to just play whatever came up on my YouTube landing page and what I'd clicked on was Rick Wakeman playing solo piano. There was a melody note that sounded janky - and this was with the two drivers sitting on the floor aimed straight up with no horns attached. That's what led me to try testing with sinewaves.
Selenium drivers were always high quality... it's why JBL bought them. A company called PRV was started by a couple of the engineers/employees that were laid off when Selenium was sold, they produced many of the same driver models but they didn't have the same quality, I suspect manufacturing was outsourced to china. The super cheap copies floating around now are probably worse again.
@conanski - For purposes of the wig-wag and sinewave tests, yes, it was straight audio all the way from the laptop audio interface. But we're also taking about power levels best expressed in microwatts.
@Cal Weldon - yes, but the JBL Seleniums only go down to 1kHz. The cab's woofers' spec sheet says their maximum usable frequency is 800Hz. The spec sheet for the updated drivers (PRV D270Ph-S) specifically recommends a second-order crossover at 500Hz which is exactly what is in the cabs now although I may yet make swap the crossover out for another that has 625Hz instead. If I can get two good and reasonably-well-matched compression drivers I'm ready to run with the configuration. With a not-terribly-precise attempt at setting the EQ and using the previous (Timpano) compression drivers, the far-field performance of the cabs while playing prerecorded music at moderate volume is quite pleasant; swapping out the 12"s made an immense difference. But the two Timpanos' responses were too different to settle for, especially because I'd be using them for synthesizers. Right now, I feel like I've got one good PRV driver and would like a second. At this point I want to just walk away from the Timpanos.
@Cal Weldon - yes, but the JBL Seleniums only go down to 1kHz. The cab's woofers' spec sheet says their maximum usable frequency is 800Hz. The spec sheet for the updated drivers (PRV D270Ph-S) specifically recommends a second-order crossover at 500Hz which is exactly what is in the cabs now although I may yet make swap the crossover out for another that has 625Hz instead. If I can get two good and reasonably-well-matched compression drivers I'm ready to run with the configuration. With a not-terribly-precise attempt at setting the EQ and using the previous (Timpano) compression drivers, the far-field performance of the cabs while playing prerecorded music at moderate volume is quite pleasant; swapping out the 12"s made an immense difference. But the two Timpanos' responses were too different to settle for, especially because I'd be using them for synthesizers. Right now, I feel like I've got one good PRV driver and would like a second. At this point I want to just walk away from the Timpanos.