EPI-like woofer-tweeter modules

[Mountainman Bob]

The specs are on the woofers' pages.
PE's DATS will do most everything but FR response(TSPs,impedence,inductance,etc).

thanks david,
must have missed the specs before......found them this time.

pro 31 10" woofer

fs, 18hz
re, 4.25 ohms
le, 0.74 mh
xmax, .25"
qm, 3.3
qe 0.37
qt, 0.33
vas, 128 ltr
90 db 1w/1m
250w rms
freq response 25-1.8k

looks like i need the DATS bundle to do the FR .....they're out of stock at the moment. would have to find a laptop....probably pawn shop is good place for that?

[dynamo]

You don’t need to buy a dats, you can build one. Instructions are here:

http://techtalk.parts-express.com/fo...ustrated-guide


For frequency response you would need a mic. Omnimic is an easy all in one setup but can be pieced together for cheaper with a mic, freeware (arta), and a blue icicle.

[Mountainman Bob]

i'm not sure if this link will work but its the FR chart for the dayton PA255-8.....what would it take to match it to the f3 @ 2.6k on the tweeter and then I also have other caps I could get the f3 on the tweets down to 2k but i'm not sure if they would handle a 1st order curve that low i'm using either the original epi style burhoe 1" concave (6ohm crosses @ 2.6k w/10uf cap) or the human 1" airspring (4 ohm crosses @2.6kk w/15uf cap).....both sound close but the nod goes to the burhoe.

what would a 1st order lpf for that pa255 look like for 2.6k and also 2k ?

appreciate it fellers

Bob

[Mountainman Bob]

ok well I suppose your thinking why should we do the the dirty work for ol' mtnman bob, so how about verifying what i've come up with?

for the pa255 if one were to use a .27 mh inductor in series and shunt a 3.9 ohm resistor in parallel just after that I come up with an low pass f3 of 2300 ......am I remotely on the right track?

then in turn I can change my cap on the tweet from a 10uf to a 12uf which will drop the hp to 2200........i'm assuming you can use the resistance value of the speaker to calculate the cap value? (not real sure)

thinking is this should lower out all those nasty distortion humps and level the impedance rise dynamo warned me about by just using an inductor straight up.

i've got these parts here in stock as close as i can get to what i need so i'm kindly stuck with these numbers for the moment.

Bob

[davidB]

It looks like you are basing your calculations on nominal values.

You need the actual impedence at a given frequency to get close to what your filter will do, or to figure where an existing filter will work.

Those tweeters have almost a third-order rolloff with just a cap(electrical+mechanical=acoustic), and the original woofers rolled off between first and second.
The Burhoe modules crossed at 1800hz with 10mf, which gives a hint as to the actual impedence in that region.

Since the frd etc files are published for the Daytons, you should try a simulation with a random tweeter, just to get a feel of what happens when you change something.
Later, if you cannot measure the tweeter, you can generate the files from the FR graph in this thread, you'll be fairly close if yours also have ferrofluid.

Sims are a great learning tool, but an actual physical book read several times(easier to go back and forth) helps generate a bigger picture.

The interweb is a wonderful thing, but, aside from errors, a random collection of ideas does not help in deciding their relative usefulness in a given situation.

[Mountainman Bob]

It looks like you are basing your calculations on nominal values.

You need the actual impedence at a given frequency to get close to what your filter will do, or to figure where an existing filter will work.

Those tweeters have almost a third-order rolloff with just a cap(electrical+mechanical=acoustic), and the original woofers rolled off between first and second.
The Burhoe modules crossed at 1800hz with 10mf, which gives a hint as to the actual impedence in that region.

Since the frd etc files are published for the Daytons, you should try a simulation with a random tweeter, just to get a feel of what happens when you change something.
Later, if you cannot measure the tweeter, you can generate the files from the FR graph in this thread, you'll be fairly close if yours also have ferrofluid.

Sims are a great learning tool, but an actual physical book read several times(easier to go back and forth) helps generate a bigger picture.

The interweb is a wonderful thing, but, aside from errors, a random collection of ideas does not help in deciding their relative usefulness in a given situation.

ok well according to the above my original burhoe tweet is 8.7 actual ohms @1800 / 6 ohm nom. / 4.7 ohm on the bench and has a 10uf cap so nothing to change there.

On the pa255 it has an impedance chart so are you saying I could just shunt a cap with a value based on the 'actual' impedance at a given freq which would decide its starting point or 'knee'

ex; at 1800hz its roughly 16 ohm......so looking at a 5.5uf cap in parallel on the woofer to get a LP f3 of 1800hz ? it can't be that simple?

[davidB]

Well, with a rising response and impedence, nothing will work quite as well as we calculate, so the fewer parts method is to increase the inductor till we get what we need.

If that's not enough to make us happy, it's on to a shunt of some kind. A sim will help you keep an eye on phase, or don't worry, try both ways.
You obviously love these speakers, so don't sweat this too much, you can solve problems when you hear them, or learn by measuring and adjusting them. It's the hobby.

[Mountainman Bob]

Well, with a rising response and impedence, nothing will work quite as well as we calculate, so the fewer parts method is to increase the inductor till we get what we need.

If that's not enough to make us happy, it's on to a shunt of some kind. A sim will help you keep an eye on phase, or don't worry, try both ways.
You obviously love these speakers, so don't sweat this too much, you can solve problems when you hear them, or learn by measuring and adjusting them. It's the hobby.

well David they say even blind squirrels find nuts occasionally.........I had a couple 6uf caps and shunted them on the woofers, nothing else;just the caps, i'll be danged if it didn't actually do what I wanted it got rid of the overlap and smoothed out the upper mid to a point thats actually quite pleasant even when ridiculously loud while still keeping it transparent and clean. calculates just under 1700 f3 though probably need a 5uf to get it around 1800

I know this might sound stupid but what exactly did I do?

Civit your crossover on top has a cap and an inductor? I thought the EPI was just a cap.....or is that inductor something you added if so how/why?

[Mountainman Bob]

aparantly after a little research what i've done is a bypass cap but without bypassing anything else.........seems to be a controversial subject, but it certainly helped my cause, the ones I used were NPE would MP be any better?

[davidB]

The voicecoil is a series inductor, it's just one that changes a bit with frequency.
Go to a Scanspeak datasheet, you will see the electrical equivalency diagram for the speaker.

Now get that amp fixed and turn down the bass a trifle, if you hear a difference between controls at neutral and bypassed something is wrong, and it's not the design of the preamp section.

Some amps will freak out if the voicecoil goes open circuit, you could put a high value resistor shunt before the cap shunt, if you go lower, say 8 or 10 ohms you can lower the woofer level a little, might be what you need for the moment, but you risk being drawn into the dark world of making your drivers and amps happy.

[Mountainman Bob]

It's a new amp since about a wk ago......all the testing i've been doing has been direct mode.
The difference putting that shunt cap in was night and day better.
Not knowing what I actually did is kindly annoying....definately need to get some testing equipment!

bob

[davidB]

New parts are always suspect. Work those wipers and switches.
if you detect anything, work them some more, you should be able to clean those unused contacts with a little use.
You don't know how long a part sits before being sold and used in manufacture.

[Mountainman Bob]

the amp is happy, if anything its happier..its running warm now after the cap shunt .....before it was rather cool for a class A/B amp. I was just wondering if maybe a higher quality cap closer to what I calculated might be better.....the 6uf 's that i used were 30 yr old NPE....guess i'll just order up a couple.

[davidB]

No. It's not happier.
If you notice heat like that your impedence is dropping too low in the shared range.
If you're buying parts put a 25-33 ohm resistor inboard of the cap.
That way when you burn the coil you'll still have an amp.

If you add the little coil you have in front the caps will see and shunt less, and your impedence will stay higher, and your amp will run cooler.

[Mountainman Bob]

No. It's not happier.
If you notice heat like that your impedence is dropping too low in the shared range.
If you're buying parts put a 25-33 ohm resistor inboard of the cap.
That way when you burn the coil you'll still have an amp.

If you add the little coil you have in front the caps will see and shunt less, and your impedence will stay higher, and your amp will run cooler.

yah I don't think your following.....it was running cooler than a class D !
Cool running class A/B amp is lifeless......whatever I did is making it run WARM (not hot) like it should be at the levels i've been pushing it to.