XO help wanted.
 

Hi. I've been reading here a while but not posting so much. It's mostly
above me but I'm trying to learn....

I have a problem, forgive me if this is the wrong forum in which to ask for
help. However I feel that I'm quite a way away from being able to sort it
out myself (although I do have a budget LCR meter that just arrived from
Hong Kong).

I have some small two-way speakers that I'm quite partial to. I don't run
them full range however, having a Jamo 4th-order isobarik passive sub, 8 ohm
nominal (with a large inductor in series with each driver and a 72uF / 50V
capacitor in series with the 'satellites out' terminals. I guesstimate that
it's passing around 210Hz and above? From the sound of it it's going down to
a bit below 100Hz, perhaps as low as 80Hz. I have a 12" Klipsch powered sub
set to 70Hz and lower, after experiementing, that seems about right. (I
don't like the Klipsch running much higher as I find it sounds a bit 'boomy'
approaching 100Hz, and above.)

The 2-ways have always sounded a bit 'confused' at the high-mid to low high
range (if you'll excuse the terminology). I don't notice it often but when I
do it bugs me. Recently I had cause to pull the XOs out (connected to the
terminal block) and I see that the 4" 'woofer' is being run full-range.
There was a 3.3uF electro cap in series with the tweeter (which I replaced
with a polyprop unit, the XO PCB was stenciled and drilled for both) there
is also a fine-wired, air-core inductor in parallel with it.

I would very much like to put a simple low-pass filter on the woofer, with
the cut-off point the same (or as close as possible) as that of the
high-pass circuit on the tweeter. However, other than the woofer being 3.6
Ohm and the tweeter being 8 Ohm (boxes rated at 4 Ohm) I know nothing about
them. They're small Philips units, quasi-ribbon tweeters, well finished with
a good rose-wood veneer and I bought them at auction.

Any and all help appreciated. TIA,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)

XO help wanted.
 

In article , ~misfit~
wrote:
Hi. I've been reading here a while but not posting so much. It's mostly
above me but I'm trying to learn....

I have a problem, forgive me if this is the wrong forum in which to ask
for help. However I feel that I'm quite a way away from being able to
sort it out myself (although I do have a budget LCR meter that just
arrived from Hong Kong).


I would very much like to put a simple low-pass filter on the woofer,
with the cut-off point the same (or as close as possible) as that of
the high-pass circuit on the tweeter. However, other than the woofer
being 3.6 Ohm and the tweeter being 8 Ohm (boxes rated at 4 Ohm) I know
nothing about them. They're small Philips units, quasi-ribbon tweeters,
well finished with a good rose-wood veneer and I bought them at auction.

Ideally, you'd equip yourself and be able to do some basic measurements.

But you could experiment 'by ear' if willing to spend some time doing so.
What I can't say is if this would be worth the effort instead of replacing
the speakers entirely. Experiments may not yield a satisfactory result if
the speaker units aren't good enough to be 'improved' by such tweaks. Or if
a simple crossover isn't up to the job. Lacking measurements or experience,
trying what you propose may involve a lot of trial and error without ending
up getting a good return!

Does you have a meter that gives a reliable AC voltage measurement that
doesn't vary much with frequency across the band from about 50Hz up to a
few kHz? If so, it could help you experiment. (I'm not sure if your "LCR
meter" also measures voltages.)

The problem with an LCR meter is that it may just do this at one frequency,
and not give a useful results for L/C/R for something that has complex
behaviour. Going that route you may really need to be able to check the
impedance at various frequencies.

Assuming you can measure AC voltages...

To test the meter, you could connect it as a voltmeter to measure the
output of your power amp and play some test tones from a CD, say, see if
the reading for the voltage stays steady as you change frequency. However
note that a cheap multimeter may only give sensible AC voltage readings for
50-60Hz.

By ear or by meter, the simple way to experiment is to try putting
something like an inductor in series with the woofer. Lacking measurements
you'd have to guess what may be the right value for inductance. Then listen
(and/or measure the effect on the ac voltage at the woofer) to decide what
effect it has had. Use that to decide if a bigger or smaller inductance
might be worth trying. Repeat until happy or exhausted. :-)

Lacking measurements you can 'guess' the speaker impedance is around 10
Ohms, then choose an inductor value on that basis. Chances are, the guess
will be wrong and the actual impedance is higher. However you can then
cheaply try connecting a resistor *across* the woofer terminals to pull
down the impedance at the roll-away point. Then experiment with resistors -
cheaper than trying different inductors. But avoid choosing a value that is
too low as the amp may not like the loading.

For real use you'd need an inductor that can handle high enough currents
and with a very low series resistance. But to save cost you could
experiment first with cheaper low-current inductors. Or even 'wind your
own'.

Beyond that, I suspect others can give you more detailed advice, and may
have experience of speaking building, etc.

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

XO help wanted.
 

"~misfit~"


** In the world of electronics, "XO" = crystal oscillator.

In the world of loudspeakers, the term "crossover" is often shortened to
X-over.

On internet forums, just about anything goes...



..... Phil

XO help wanted.
 

Somewhere on teh intarwebs Phil Allison wrote:
"~misfit~"


** In the world of electronics, "XO" = crystal oscillator.

In the world of loudspeakers, the term "crossover" is often shortened
to X-over.

Thanks Phil, I believe I mentioned that I was on the low-end of the
learning-curve? ;-)

On internet forums, just about anything goes...

Indeed. However it's easiest if we're all speaking the same language yes?

Heh, your surname remined me that I must get around to doing *something*
with the boxes containing all 12 drivers from a set of Allison Ones (January
1978) that I have in the back room. They've been there for years and I never
did get around to doing anything with them.

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)

XO help wanted.
 

Somewhere on teh intarwebs Jim Lesurf wrote:
In article , ~misfit~
wrote:
Hi. I've been reading here a while but not posting so much. It's
mostly above me but I'm trying to learn....

I have a problem, forgive me if this is the wrong forum in which to
ask for help. However I feel that I'm quite a way away from being
able to sort it out myself (although I do have a budget LCR meter
that just arrived from Hong Kong).


I would very much like to put a simple low-pass filter on the woofer,
with the cut-off point the same (or as close as possible) as that of
the high-pass circuit on the tweeter. However, other than the woofer
being 3.6 Ohm and the tweeter being 8 Ohm (boxes rated at 4 Ohm) I
know nothing about them. They're small Philips units, quasi-ribbon
tweeters, well finished with a good rose-wood veneer and I bought
them at auction.

Ideally, you'd equip yourself and be able to do some basic
measurements.

Indeed. I'm trying to find downloads of any good freeware that I can use.
However, circumstances have left me an invalid, having lost my home and
business, with only welfare as an income, and a few remnants of my audio
gear. So, as I have little to no spending money I'm trying to get the best
listening experience that I can from what I have. Hence the question and the
attempt at learning as much as I can.

Sadly opioid painkillers rob me of a large chunk of my ability to learn, to
form memories. I used to learn as easy as pie. I graduated two years younger
than my class-mates... Now I feel like a dunce sometimes. :-/

But you could experiment 'by ear' if willing to spend some time doing
so. What I can't say is if this would be worth the effort instead of
replacing the speakers entirely.

Not really an option, see above.

Experiments may not yield a
satisfactory result if the speaker units aren't good enough to be
'improved' by such tweaks. Or if a simple crossover isn't up to the
job. Lacking measurements or experience, trying what you propose may
involve a lot of trial and error without ending up getting a good
return!

Frankly I was rather hoping there was maybe a software tool that I could
enter the values of what's already in the boxes, the drivers and high-pass
circuit and get some idea of where the high-pass point is and what
components I would need to build a low-pass for the woofer that works at the
same point.

Does you have a meter that gives a reliable AC voltage measurement
that doesn't vary much with frequency across the band from about 50Hz
up to a few kHz? If so, it could help you experiment. (I'm not sure
if your "LCR meter" also measures voltages.)

I have a fairly cheap DMM other than the LCR meter which measures AC but
have never discovered if it does what you ask, nor do I know how to test it
to find out. All I know about it is that it cost $30 and is rated for mains
electricity work. It seems quite good, with a 32-position rotary dial in the
middle, AC range has 5 positions from 200mV to 750V.

The problem with an LCR meter is that it may just do this at one
frequency, and not give a useful results for L/C/R for something that
has complex behaviour. Going that route you may really need to be
able to check the impedance at various frequencies.

I mainly bought the LCR meter so that I could measure inductors and
capacitors so that I might a) re-use components that I have from older
systems or b) as you mention below, maybe hand-wind my own inductors. My
existing DMM didn't measure inductance and has a lower range of capacitance.

Assuming you can measure AC voltages...

To test the meter, you could connect it as a voltmeter to measure the
output of your power amp and play some test tones from a CD, say, see
if the reading for the voltage stays steady as you change frequency.
However note that a cheap multimeter may only give sensible AC
voltage readings for 50-60Hz.

Ok, thanks. Seems I'll be searching for a download of either a test-tone CD
or, preferably, software that I can run on my second laptop that will output
variable tones.

By ear or by meter, the simple way to experiment is to try putting
something like an inductor in series with the woofer. Lacking
measurements you'd have to guess what may be the right value for
inductance. Then listen (and/or measure the effect on the ac voltage
at the woofer) to decide what effect it has had. Use that to decide
if a bigger or smaller inductance might be worth trying. Repeat until
happy or exhausted. :-)

Yes, that sounds to be roughly what I tought I might have to do. As they
are, the X-over is built onto a small PCB connected to the speaker terminal
block. I don't want to keep pulling the things to bits so might open them
and run the wires for the woofer out the reflex port. That way, if I do get
exhausted (a real possibility with my back pain) I can simply wire them back
to the terminals along with the amp output until next time. :-)

Lacking measurements you can 'guess' the speaker impedance is around
10 Ohms, then choose an inductor value on that basis.

As I mentioned, the woofer is in fact rated 3.6 Ohms. At least that's what's
written on it, with 8 Ohms written on the ribbons, and the sticker on the
back says nom. imp. 4 Ohms.

Chances are,
the guess will be wrong and the actual impedance is higher. However
you can then cheaply try connecting a resistor *across* the woofer
terminals to pull down the impedance at the roll-away point. Then
experiment with resistors - cheaper than trying different inductors.
But avoid choosing a value that is too low as the amp may not like
the loading.

Ok, thanks. I hope that the amp's protection circuitry's working fine as, if
it's not and it's needed I'm not in a position to replace it.

For real use you'd need an inductor that can handle high enough
currents and with a very low series resistance. But to save cost you
could experiment first with cheaper low-current inductors. Or even
'wind your own'.

I've been grabbing inductors out of scrapped computer / server
power-supplies lately (so many I didn't...). They're mostly toroidial and I
didn't think they'd be of use (other than a source of wire) but then I read
about a speaker company using toroidial inductors on their woofers, like
it's a great thing... The trouble is the highest indusctance I have is only
0.4mH but it's 1.3mm diameter wire.

Beyond that, I suspect others can give you more detailed advice, and
may have experience of speaking building, etc.

I was hoping so, hence holding off on replying. However it seems not.

Thanks Jim. I'd pour you a dram or two of of Uigeadial if only I still had
some...
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)

XO help wanted.
 

"~misfit~"
Phil Allison wrote:

** In the world of electronics, "XO" = crystal oscillator.

In the world of loudspeakers, the term "crossover" is often shortened
to X-over.

Thanks Phil, I believe I mentioned that I was on the low-end of the
learning-curve? ;-)

On internet forums, just about anything goes...

Indeed. However it's easiest if we're all speaking the same language yes?

Heh, your surname remined me that I must get around to doing *something*
with the boxes containing all 12 drivers from a set of Allison Ones
(January 1978) that I have in the back room.


** I always preferred the name sake connection with Allison aero engines.

My first ride in a plane was in a Lockheed L188 Electra with four, Allison
T56 turboprops.

http://en.wikipedia.org/wiki/Allison_Engine_Company



..... Phil

XO help wanted.
 

In article , ~misfit~
wrote:
Somewhere on teh intarwebs Jim Lesurf wrote:
In article , ~misfit~


Ideally, you'd equip yourself and be able to do some basic
measurements.

Indeed. I'm trying to find downloads of any good freeware that I can
use.

I'll comment in more detail later if I get a chance. But you should be able
to find various freeware around. If you use Linux (or RISC OS) you can use
some of the programs on http://www.audiomisc.co.uk/software/index.html
to do things like create test files/audio CDs and analyse the results in
various ways. However I suspect the main challenge will be carrying out
suitable measurements to get the data needed to decide what kind of
crossover may be useful.

FWIW I've never bothered to make or mod my own speakers. I've always
regarded it as being too much bother as it can be very hard to reach a good
result. Although I've designed or fiddled with much of the electronics in
my audio systems, I tend to leave speakers to the professionals as I regard
that as the 'hard part' of home audio. :-)

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

XO help wanted.
 

Somewhere on teh intarwebs Jim Lesurf wrote:
In article , ~misfit~
wrote:
Somewhere on teh intarwebs Jim Lesurf wrote:
In article , ~misfit~

Ideally, you'd equip yourself and be able to do some basic
measurements.

Indeed. I'm trying to find downloads of any good freeware that I can
use.

I'll comment in more detail later if I get a chance. But you should
be able to find various freeware around. If you use Linux (or RISC
OS) you can use some of the programs on
http://www.audiomisc.co.uk/software/index.html
to do things like create test files/audio CDs and analyse the results
in various ways.

Alas, I use Windows, and 32-bit XP Pro at that! I have a really nice little
IBM X32 ThinkPad that I'm going to use for my audio / electrical stuff. I've
installed a mate's (older) copy of BassBox Pro and it comes with a programme
called X-Over [3] Pro. However so far I've only modelled a couple of speaker
boxes, I haven't touched the X-Over part. It's supposed to be quite easy to
learn but a side-effect of my meds is bad short-term memory so learning
*anything* is extraordinarily difficult! I suppose that being 50 y/o doesn't
help either.

Annoying as, at school I was a social outcast due to being put ahead two
years and *still* coming top of the class in most subjects. :( I find the
contrast alarming.

However I suspect the main challenge will be
carrying out suitable measurements to get the data needed to decide
what kind of crossover may be useful.

My cheap Chinese LCR meter seems to be pretty good in that regard, at least
for inductance. Well, the only reason I say that is that it's repeatable
and, in X-Overs I've pulled apart from old speakers the parts from both
speakers measure the same. This is it:
http://www.dealextreme.com/p/4070l-2...ltimeter-26768


Unfortunately you can't use the probes to measure an inductor, the ends of
the component itself have to be inserted into the sockets on the meter.
That's annoying as I have some nice speakers around and I'd love to know the
values of the components in their X-Overs. (Also I have a friend who will
let me pull the woofer from his speakers to take a peek at how well they're
made [he's curious too but wouldn't do it himself] and it would be nice to
be able to measue inductors in situ. shrug Oh well.) Caps and resistors
have the value written on them. However I'd rather not unsolder everything
to test the inductors.

FWIW I've never bothered to make or mod my own speakers. I've always
regarded it as being too much bother as it can be very hard to reach
a good result. Although I've designed or fiddled with much of the
electronics in my audio systems, I tend to leave speakers to the
professionals as I regard that as the 'hard part' of home audio. :-)

Heh! Yes, hard and expensive. I made my first set of speakers over 30 years
ago as a teenager as I couldn't afford the prices decent boxes cost. However
I didn't bother too much with X-Overs back then (similar to a lot of
manufacturers of the time), simply adding capacitors to protect mids and
tweets. Woofers ran full-range.

Then there was a period when I was earning well and some of what I didn't
save towards starting my business I allowed myself to spend on good speakers
(and other audio gear). Alas, now I've almost come full-circle (except I
can't take part-time jobs like I used to during school holidays) and am poor
but still want to have good-sounding gear.

I find that I don't have to spend a fortune on an amplifier to get something
that sounds pretty good. It's the speakers where there's the biggest
variable, and cost. I simply can't afford to buy (built-up) anything
remotely like I can appreciate so I'm having to build my own again. However
X-over technology and understanding have come a long way and also my ears
have become more discerning.

So, due to financial constraints I'm building my own speakers, mostly using
scavenged drivers (so I don't always coughevercough have the T/S
parameters for them). Sometimes I can find them on teh webs which is great
but mostly I'm flying blind, other than what I can measure.

Oh, as a side-note, although I can wield a soldering-iron with the best of
them I've never learned much about electronics. My experience has mainly
been limited to replacing components (caps on PC motherboards being the big
one) and sticking wires together. I'm currently heh! reading this:
http://electronicstheory.com/COURSES...ICS/e101-1.htm and looking for
more similar educational material. I have a couple old amps that I'd love to
be able to fix and, more than anything else being an invalid with an active
mind (albeit one that has trouble remembering) I *need* to have something to
keep my mind occupied and the less it costs the better! g

If you can suggest any sites that might help me in my quest to learn (but
basically from the start) I'd love to have the URLs. Reading forums like
DIYAudio makes me wish I could build my own amplifier.... ;-)

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)

XO help wanted.
 

Somewhere on teh intarwebs Phil Allison wrote:
"~misfit~"
Phil Allison wrote:

** In the world of electronics, "XO" = crystal oscillator.

In the world of loudspeakers, the term "crossover" is often
shortened to X-over.

Thanks Phil, I believe I mentioned that I was on the low-end of the
learning-curve? ;-)

On internet forums, just about anything goes...

Indeed. However it's easiest if we're all speaking the same language
yes? Heh, your surname remined me that I must get around to doing
*something* with the boxes containing all 12 drivers from a set of
Allison Ones (January 1978) that I have in the back room.


** I always preferred the name sake connection with Allison aero
engines.
My first ride in a plane was in a Lockheed L188 Electra with four,
Allison T56 turboprops.

http://en.wikipedia.org/wiki/Allison_Engine_Company

Allison made some impressive engines, that's for sure. :-)
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)

XO help wanted.
 

In article , ~misfit~
wrote:
Somewhere on teh intarwebs Jim Lesurf wrote:


I'll comment in more detail later if I get a chance. But you should be
able to find various freeware around. If you use Linux (or RISC OS)
you can use some of the programs on
http://www.audiomisc.co.uk/software/index.html to do things like
create test files/audio CDs and analyse the results in various ways.

Alas, I use Windows, and 32-bit XP Pro at that!

In case it helps, the programs on my website all come with the full source
code in 'C'. So if you can use GCC on your windows box you could tweak and
recompile the Linux code and make your own versions. I'm quite happy for
people to do that. The programs and code are provided for anyone who wants
to make (non commercial) use of them.

[snip]

If you can suggest any sites that might help me in my quest to learn
(but basically from the start) I'd love to have the URLs. Reading
forums like DIYAudio makes me wish I could build my own amplifier....
;-)

Can't suggest websites for speaker building, I'm afraid. What little I know
about that is from collected ages of mags, journals, etc. But if you can
solder and bend metal I'd suspect you find making a decent amp far easier
than making a speaker that sounds good. People seem to have drawn to
speaker making because it seems 'easy' in terms of electronics. But there
are lots of decent amp designs around.

And if you want something like a DIY preamp there are tricks you can
employ. My favourite is to buy a second-hand Quad 34 and then modify it to
your heart's content. With the manual they are easy to fiddle about.
Everything is neatly laid out and labelled. No need to even bend metal or
make boards.

IIRC someone also used to sell upgraded boards for the venerable 303. So
there are all kinds of options.

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html