advertisement


JPW Sonata tweaking...

RichShortland

RichShortland

pfm Member
Hi all,
I've decided to keep the Sonatas I have (had) for sale and give them some love.
I think they could be proper little giant killers with some work as they have some great things going for them:
- good solid sealed cabinets
- great mid/bass unit (Elac 6.5")
- lovely midrange as a result...
- simple 6db first order crossover

Plan so far is to change the 3.3uF cap in the crossover feeding the tweeter for a decent poly one.
Then damp the cabinet internals with bitument sheet, reseal all internal edges and coat everyting with some liquid rubber. Plus add more hollow fibre fill.

But - should I thing about some other stuff too? Internal bracing? Brace/support the magnet of the mid/bass?
I'm very keen on replacing the tweeter but what with? A Monocor DT250 perhaps?

Slightly way out there was a thought of double-skinning the cabinet external sidewalls with sand loaded mdf sheets (and then prettying them up).

Any suggestions welcome :)
 
If you're going to swap an electrolytic for a poly cap, the poly will likely have lower ESR which may make the tweeter sound too hot so you'll want to compensate by adding resistance in series.

I'd have thought that the cabinets are small enough to be quite rigid and therefore bracing might not have much effect.
 
Mike P said:
If you're going to swap an electrolytic for a poly cap, the poly will likely have lower ESR which may make the tweeter sound too hot so you'll want to compensate by adding resistance in series.

I'd have thought that the cabinets are small enough to be quite rigid and therefore bracing might not have much effect.
Click to expand...

Hi Mike,
Thanks for this. The mid/bass only features an inductor. Is there any mileage in considering running this direct do you think? Or not worth it?
 
RichShortland said:
Hi Mike,
Thanks for this. The mid/bass only features an inductor. Is there any mileage in considering running this direct do you think? Or not worth it?
Click to expand...
Personally I wouldn't but I'm no expert.

If you can, measure the ESR of the old electrolytic that's in series with the tweeter and also that of the new poly cap and calculate the difference. Then increase the value of the series resistor by that amount.
 
From what I remember from reading of the development of EPOS ES14, both the bass/mid cone material and driver suspension need be highly self-damping in order for the driver to roll-off in a somewhat controlled manner on its own. Although Sonata has a nice quality unit, paper isn't likely to be up to the job.

As to changing tweeters, although the little Audax TW010xx are difficult to criticize when not stretched too far, there are some considerations to keep in mind when changing these out for something like the Monocor DT250.
For example, the Audax units resonance frequency (Fs) is 3kHz, above which it doesn't begin to level off until circa 4Hz. Even then, Audax's own suggested application is 1st or 2nd order at Fc 6kHz, using 3uF or 3.3uF (the latter with a 0.12mH inductor in parallel for 2nd order) to achieve nominal power handling of 25 or 45W respectively. Therefore, changing the Audax out for the Monocor DT250, will require more than simply a possible resistor change for level match, as the Fs is 1.7kHz here, with a recommended Fc of 3kHz (12dB/oct, i.e. a cap in series plus an inductor in parallel) in order to meet the power handling spec of 50W RMS. No doubt, the parallel inductor across the tweeter could be omitted, perhaps compensating for that change by upping the Fc via a higher value cap, so as not to push one's luck wrt power handling with the shallower 6dB/octave slope.

All of the above leave the existing inductor that is in series with the bass/mid driver causing a greater overlap of frequency bandwidths, something that, from experience, often makes for a much worse net result than keeping an existing 'lesser' tweeter in. For reference, 1st order Butterworth (6dB/oct) with 8Ohm tweeter/8Ohm bass has C1 = 5.7uF and L1 = 0.36 mH (in series, tweeter and woofer respectively).

JPW Sonata crossover: 3.3 uF, 50v Elcap with 4.7Ohm resistor. Likely a 0.21mH inductor in series with the bass/mid, according to Butterworth 6dB/oct with Fc @ 6kHz.
main.php
 
I think changing the capacitor is a good idea.
Be very careful about changing anything on the cabinet! Only add things that you can remove later if you don't like the effect!
At the minute the Sonota is probably lively, fun. All the cabinet mods you suggested will make it nominally "better" but possibly not so interesting to listen to.
Try clamping a couple of additional side panels, cut to size, firmly to the existing cabinet. This will have 2 or more effects, the side panels resonances will change (better/worse?) and the efficiency of the woofer will go up very slightly between about 100 and 500Hz (baffle step issue) giving you a different tonal balance.

Remember we don't all want the same things from hifi, some people want "live" type sound, probably full of colourations,and other are interested in hearing every last piece of detail in a recording.
 
I wouldn't remove the inductor. It's there to form the baffle step correction and low pass filter.

Adding more stuffing will lower Qtc and as a result give you less bass. You could experiment here but I'd be surprised if JPW didn't get this right in the first place.

Adding material to the outside of the cabinet could affect the crossover negatively, but It's hard to say what will happen exactly without measurements.

As Mike P said, you could replace the electro cap for a poly but you may need to increase the value of the resistor because of the reduced ESR (usually around an ohm will do).

Bitumen sheet may help reduce cabinet vibrations, so that could be worth a try.

Unless you're very lucky, you'd need to redesign the crossover if you change the tweeter. To do it properly you'd need measuring gear and knowledge of crossover design. Online crossover calculators don't work...

Suggestions.

Build a cabinet with the same front baffle dimensions and keep the driver positions exactly the same but build the cabinet with constrained-layer-damping (CLD). That works much better than adding bitumen sheets to the walls. Read the Tannoy thread to give you some ideas - LINK
 
When we were trying to develop a small speaker in the 80's we tried a prototype which was a small MDF box, lined with latex floor levelling compound which was angled by tilting the box as it set, so there were no absolutely parallel internal faces, and quite high mass and good damping. We then fitted that box inside a rather nice solid wood box (prototype was pine, but the intent was to offer more exotic finishes) separated by a thin (about 7mm) layer of foam - we tried several. It worked well, with nothing escaping through the cabinet walls, but it ended up rather larger than we wanted for the internal volume we needed and the estimated cost and speed of production was completely non-viable for a commercial product. Sounded good, and I wish that was the prototype that I kept when we abandoned the project. So making an external lining box could work, and we found that the FR was not really affected, but I was slightly uneasy about the imaging in comparison with the narrower version, though my colleagues weren't...
 
ADDENDUM: A further point I meant to make above, before nodding off, is that, if you are to substitute a posher 1" metal dome, then it would be a shame not to take advantage of its superior resolution down to lower Fq than the cheap Audax can manage.

IOWs, with the existing Audax, the bass/mid is expected to do well at higher frequencies than usual. A side benefit of this is that it also places the crossover point further above the listener's most sensitive/critical hearing range, by necessity - which I believe is why so many small driver 2-way speakers do so well with this unit (including Royd's relatively expensive last model to use it - RR1). The negative bit is that these sorts tend to go off in all directions sooner as volume increases.

This is why I brought up the subject of potentially changing out L1 for something that would compliment a higher value C1 that places the crossover point down nearer to the recommended 3kHz Fc of DT250.

You really wouldn't want to listen to the bass/mid attempting the same range due to overlap.

Certainly, leaving the existing L1 in place and level matching DT250 with an alternative resistor as necessary is an option, however, I doubt that it would be worth the woodworking bother vs. fresh poly caps.

I would like to know what bits make up the crossover of JPW Sonata Plus, though...
69ee7223f63458cd80749ccf07d815ce.jpg
 
Hello all,
Thanks very much for the detailed feedback and suggestions - it's been very interesting digesting all the above.

Sit rep: I resealed/damped the internal cabinet walls today which was nice and easy, plus has certainly added weight and 'deadenned' the noise from a knuckle wrap on the top or sides...
I then swapped out the cap for a nice poly, plus the resistor for a higher value as suggested.
I also added some damping material to the back of the tweeter, plus the magnet of the bass driver.
I damped the new cap with a few layers of heat shrink too.

All back together and gave it a listen...
Sounded very nice, smoother, less perceived sibilance coming from the tweeter, bass felt nice and tight as before and mid range was good.
However - when I turned up the wick a little the coherence was lost and the bass turned very one-note and thumpy.
When I stripped the cabs I was very surprised by the small amount of wadding (only really around the sides/top/bottom).
So I did wonder if when turned up I was was hearing the rearward sound of the cone reflecting off the back of the internals and coming straight out of the paper cone woofer... This did not feel right.

So, I whipped out the bass units and began experimenting with more wadding.
Tried a bit more, then more and finally quite a lot more and bingo - a much more grown up, sophisticated, detailed and coherent presentation, even when turned up a bit.

I do accept this is subjective, and taking some of the points above, I have certainly changed the sound of the speakers. Is it better? To my ears yes, but I've added more of what I consider 'better' which might not be the same for you / others.

A question - will the cap need 'running in'?

Thanks, R
 
No the film cap won't need running in but it'll take you a little while to get used to it.

Your modifications are quite similar to what I did with my Diamonds and I've been really pleased with how they turned out.
 
Mike P said:
No the film cap won't need running in but it'll take you a little while to get used to it.

Your modifications are quite similar to what I did with my Diamonds and I've been really pleased with how they turned out.
Click to expand...

I am really pleased so far too - they are in a room far too big for them and not quite perfectly positioned but sounding good nonetheless. I think with the right sub woofer they could be VERY good...
 
RichShortland said:
I am really pleased so far too - they are in a room far too big for them and not quite perfectly positioned but sounding good nonetheless. I think with the right sub woofer they could be VERY good...
Click to expand...

So... with a little matching JPW sub I found they are very nice!..
Just still not as good as my usual set of main speakers (which was always going to be a tough ask...) so they and the sub are now up for sale in the classifieds :)
 
I bought two pairs (in fact three, one pair for the kitchen!) and a pair of Gold monitors, all JPW. The idea was to build a loudspeaker with as little compromise on quality, although inexpensive to produce. I settled on the Sonatas, two pairs, mounted sideways and one on top of the other, but with another cab (Gold Monitor without drive units) in between. The idea of the extra cab is to mount a Fountek CD3 ribbon tweeter. Here:

mdr v4

So I tried a mock-up, with just the two Sonatas, above and below the third cab. Initially, just a little disappointed, until I placed some EVA foam (gym floor tiles) between all cabs. Much better.

What I will be doing: looking at the box walls, external only, adding materials which will damp without adding too much mass. Add internal wadding (choice of many sheep/alpaca wools, mixes and rockwool). Add the Fountek tweeter (or super tweeter) xover at 4.5kHz.

So far, I am delighted with the thing. I had it as one channel (fed by a stereo class D amplifier) against an integrated stereo amplifier feeding one Usher 604 floorstander. Although the JPW hadn't the bass 'speaker connected, I preferred that channel. I was listening to Sibelius 5th Symphony, where the ending has several single chords separated by near silence, except for ambience. When the ambience came, the sound swung from the middle to the Sonata 'speakers, as they reproduced the decay properly, the Ushers were far too damped.

I will continue with their development, looking at every aspect of their design.
 
Active could be fun to have a mess around with, especially with a better/different tweeter. Easy(er) to adjust crossover frequency and relative level of tweeter to woofer. They are not exactly sophisticated in crossover etc in the first place (no compensation for any peaks or resonances in drivers, no time alignment etc) plus a new tweeter means you are in uncharted territory anyway. Some cheap but good gain clone or class D modules and away you go... or even just a spare stereo amp that happens to be around.
 
You must log in or register to reply here.


advertisement


Back
Top