Most electrolytic caps have their tolerance on the positive side of the nominal stated value. Its almost not worth installing electrolytic caps instead of metalized film as the price for decent quality electrolytics are closer to film caps. The exception is when some loss is desired, such as in LP filter applications parallel across woofers. Some dampening is usually preferred via increased series loss to soften some of the filter peaking. It can also be accomplished when using film/foil caps with added series resistance. However, I've discovered that sometimes the combination of the resistance and capacitance together with the residual inductance of the film cap windings can induce a multiple frequency peaking response all along the rolloff slope of the driver. This is especially the case when the driving amplifier has very low output impedance and the driver is still showing extended output above the chosen filter cutoff. Higher Le drivers with less BL aren't so much affected by this.
Thanks. In this crossover, all inline/high pass caps are polys, and electrolytics are used for low pass in parallel with the woofer (3) and mid (1).
So replacing with electrolytic like for like seems to be the way to go.
So replacing with electrolytic like for like seems to be the way to go.
Unfortunately in Oz we pay a premium for crossover parts: a good quality 20 microfarad film cap costs between $20 and $30, with premium caps adding more numbers or zeroes.
I don't think I ever listened to these at Encel as they were way beyond budget, but Interdyn's premium speakers had a great reputation. We later bought a pair of Interdyn 212s (8" two ways) which sounded excellent.
Geoff
I don't think I ever listened to these at Encel as they were way beyond budget, but Interdyn's premium speakers had a great reputation. We later bought a pair of Interdyn 212s (8" two ways) which sounded excellent.
Geoff
I actually like the design of this crossover > simple yet thoughtful, especially the two resistors in the midrange circuit and their placement.
With this arrangement the two resistors realize FOUR functions >
1. The 2.2 ohm provides a minimum impedance floor.
2. The 2.2 ohm + 15 ohm together create a level reduction for flat response.
3. The 15 ohm alone fine-tunes the crossover frequency.
4. The 15 ohm provides a nice subtle damping across the midrange driver.
I'm curious about the cap values across the mid and woofer though, they numbers are big which would imply quite a low, low pass point in each case. I know it's pure speculation without confirming the inductor values, so I'll wait until my new LCR meter turns up...
YES > I was just having another look at XO, and the cap. across the midrange looks VERY high value given the tweeter is about 5Kz.
20uF across the mid. would normally give you a F of around 700Hz @ 8 ohms (second order). This value is really quite odd 😕
I think finding-out the inductor values will go a long way as to explain things.
It is possible that your speakers were designed to have a particular 'sound' and to not actually be flat > you know, sort of 'BBC Sound'.
20uF across the mid. would normally give you a F of around 700Hz @ 8 ohms (second order). This value is really quite odd 😕
I think finding-out the inductor values will go a long way as to explain things.
It is possible that your speakers were designed to have a particular 'sound' and to not actually be flat > you know, sort of 'BBC Sound'.
Yes, all drivers are marked as 8 Ohms, including the tweeter (SEAS are commonly 6 Ohms).
I was wondering if maybe the resistors are doing something there...
I was wondering if maybe the resistors are doing something there...
Still interestingly strange > the 15 ohm resistor across the mid. would need to be something like 5 ohms to explain the 20uF cap.
I am now thinking that the second inductor of the midrange circuit, the one across the mid. driver is probably a significantly low value >
and hence the 2.2 ohm resistor used to keep minimum impedance to a reasonable level.
I am now thinking that the second inductor of the midrange circuit, the one across the mid. driver is probably a significantly low value >
and hence the 2.2 ohm resistor used to keep minimum impedance to a reasonable level.
Last edited:
I can’t measure ESR until I get my new meter, but the 20uF ELCAPS measured at 27uF, and the 30uF measured at 57uF!
I ended up doing parallel Jantzen Electrolytics to get exactly measured 20uF (a 15 and 3.3) and 30uF (a 22 and 4.7), given even the new ones are 10% over.
I got a reply from Aranmar Acoustics - apparently this model pre-dates their involvement in Interdyn speakers so the only way I’ll be able to find the inductor values is to measure them.
They have come up very well, and sound pretty good with the replacement caps anyway.
They have come up very well, and sound pretty good with the replacement caps anyway.
I got my new test tool. The inductors measure:
2.2mH/1.5 Ohm for the mid series,
0.04mH/1.1 Ohm for the mid parallel
2.4mH/1.5 Ohm for the tweeter series.
So you were right about it being a very low value….
At least things now seem to make sense, and if you like the sound, it's a job well done 🙂
It would indicate that as a 'stand alone driver', the midrange unit is quite a high efficiency driver.
I'm tempted to 'mock-up' the midrange circuit and see what filtering it provides the driver (using an 8 ohm resistor).
If I do, I'll post the graph.
It would indicate that as a 'stand alone driver', the midrange unit is quite a high efficiency driver.
I'm tempted to 'mock-up' the midrange circuit and see what filtering it provides the driver (using an 8 ohm resistor).
If I do, I'll post the graph.
I found a tool to calculate filter values but it didn't consider the load resistance, essentially:
3.3uF = 6kHz (Tweeter HP)
20uF + 0.04mH = 5.6 kHz (Mid LP)
20uF + 2.2mH = 758 Hz (Mid HP)
40uF + 2.4mH = 513 Hz (Woofer LP)
So that seems reasonable, plus what all the resistors including the one in series with 30uF across the woofer
3.3uF = 6kHz (Tweeter HP)
20uF + 0.04mH = 5.6 kHz (Mid LP)
20uF + 2.2mH = 758 Hz (Mid HP)
40uF + 2.4mH = 513 Hz (Woofer LP)
So that seems reasonable, plus what all the resistors including the one in series with 30uF across the woofer