High-end 3-way active Avalon look-alike

[jleaman]

Looks so nice. I like & impressed

 

[fgroen]

Thanks guys! I'm blushing.

Later more on the electronics side.

Cheers, Fokke.

 

[cloudior]

fgroen -> Great job!

Where did you get this bird's eye maple vineer?

Regards.

 

[fgroen]

fgroen -> Great job!

Where did you get this bird's eye maple vineer?

Regards.

I ran into this vineer at my local speakers supplyer. He still had a couple of sheets of Lignaflex on stock. I bet Lignaflex (or similar paper backed vineer) can be purchased worldwide. Look for local supplyers, since sending the stuff by mail is a bit awkward.

 

[fgroen]

They're getting there!

Breaking in is still in progress, but since a week, there's already a significant difference since the first notes. Also a couple of things have been done.

I was puzzling with setting up the loudness levels of the individual drivers, by ear. This resulted in "nice sounding speakers", but lacking reality. Some recordings would sound OK, others too bass heavy. So I was in the process to turning back bass to taste, but that was not the way.

Then I decided to look up the specs of the drivers (all German made, so I expected documentation to be immaculately kept. the sensitivity spec of the C2-90 hinted to this already: 93,15 dB. )

So I set up an excel sheet to easily calculate voltage levels of the filter outputs given the sensitivities of the drivers. Setting up the filters with the obtained results was 10 minutes work. It turned out that bass was too loud (had to be reduced as well as treble).

That done, the whole picture fell into place. Bass had disappeared, treble was no longer there, no more midranges. Just music. I lost track of time that night. CD after CD disappeared in the drawer of the player and every time I was staggered what information was hidden on my CD's. Information I never heard before. I thought I knew my CD collection, but I can start over again to discover what's hidden on all those discs. Amazing.

Also at low levels the listening becomes interesting and detailed. Clearly the midrange is breaking loose. Detail starts to be staggering. A kind of paradox presents itself: the speakers sound really calm and easy. Nevertheless all small detals can be heard, and proportionality between all sonic events in the soundstage is very realistic. That makes, that although there is a high level of detail, there is no hint of fatigue. Listening at a realistic level can be done for hours without any trouble.

Today I ordered new caps for my filter, to alter the range of crossover frequencies between woofer and midrange. Currently, the range is 66 to 257 Hz, and will be 208 to 807 Hz. At the moment, I'm playing at 257 Hz, and that is still slightly too low for the C2-90. I think that 350-450 Hz will be optimal for the C2-90.

 

[Jussi]

I was puzzling with setting up the loudness levels of the individual drivers, by ear. This resulted in "nice sounding speakers", but lacking reality. Some recordings would sound OK, others too bass heavy. So I was in the process to turning back bass to taste, but that was not the way.

You can't measure them?

That done, the whole picture fell into place. Bass had disappeared, treble was no longer there, no more midranges. Just music. I lost track of time that night. CD after CD disappeared in the drawer of the player and every time I was staggered what information was hidden on my CD's. Information I never heard before. I thought I knew my CD collection, but I can start over again to discover what's hidden on all those discs. Amazing.

Also at low levels the listening becomes interesting and detailed. Clearly the midrange is breaking loose. Detail starts to be staggering. A kind of paradox presents itself: the speakers sound really calm and easy. Nevertheless all small detals can be heard, and proportionality between all sonic events in the soundstage is very realistic. That makes, that although there is a high level of detail, there is no hint of fatigue. Listening at a realistic level can be done for hours without any trouble.

Sounds like mission accomplished?

Today I ordered new caps for my filter, to alter the range of crossover frequencies between woofer and midrange. Currently, the range is 66 to 257 Hz, and will be 208 to 807 Hz. At the moment, I'm playing at 257 Hz, and that is still slightly too low for the C2-90. I think that 350-450 Hz will be optimal for the C2-90. [/B]

This part I didn't get. 66 to 257Hz and 208 to 807Hz? What do you mean? I understanded 257Hz as the present xover frequency between mid and woofer. Why do you think thats too low for the C2-90?

Jussi

 

[fgroen]

You can't measure them?

No. I don't have a measuring set-up at present. Still working on that though. It's an indispensable tool to do serious tuning.

This part I didn't get. 66 to 257Hz and 208 to 807Hz? What do you mean?

Sorry, I think I was a bit short on additional information here.

The THEL filter is adjustable with 16-position DIP switches. With these switches, the user selects a resistor value from a resistor bank which, combined with the value of the capacitor, will give a certain roll off frequency. Since this is a 4th order filter, there are 4 DIP switches and 4 C's per channel.

This set-up works more accurately than a multi-potmeter setup, where there is always a difference of resistance values between the potmeters.

Drawback of only 16 preselectable frequencies is, that either the range is large and the selectability is very coarse, or vice versa. THEL sets up their filters with a 2-3 octave range to have sufficient selectivity. With a set of C's, a specific range can be selected.

An externally hosted image should be here but it was not working when we last tested it.

Left column is switch setting, column right of that is the resistor value that is configured with that switch setting. Top row is the chosen capacitor value. At the moment I have 47 nF in my filters, and I ordered 15 nF.

I understanded 257Hz as the present xover frequency between mid and woofer. Why do you think thats too low for the C2-90?

The C2-90 starts to seriously distort below 200 Hz. In order to have sufficient attenuation at 200 Hz, a crossover frequency of approx. one octave higher is recommendable. (400 Hz crossover would mean a -24 dB output at 200 Hz for the C2-90) I estimate that currently the level at 200 Hz is approximately -8 dB.

Furthermore, putting the C2-90 in a closed box enclosure of 3-4 litres will cause the C2-90 to gently roll off below 300Hz, causing a slight dip (+/- -1dB) in total response.

So... Mission not entirely accomplisched yet, but we're getting there.

 

[Jussi]

The THEL filter is adjustable with 16-position DIP switches. With these switches, the user selects a resistor value from a resistor bank which, combined with the value of the capacitor, will give a certain roll off frequency. Since this is a 4th order filter, there are 4 DIP switches and 4 C's per channel.

Ok. Now it makes sense.

The C2-90 starts to seriously distort below 200 Hz. In order to have sufficient attenuation at 200 Hz, a crossover frequency of approx. one octave higher is recommendable. (400 Hz crossover would mean a -24 dB output at 200 Hz for the C2-90) I estimate that currently the level at 200 Hz is approximately -8 dB.

Yep. At least excursion is rising and therefore causing more 2nd harmonic. I just thought 450Hz is a bit high for a 7" midrange. I had C2-77 Accutons some time ago and they performed well with 250Hz 4th order highpass. That was adequate to kill the distorsion rise at lower frequences. At least up to 95dB SPL in anechoic room.

Jussi

 

[fgroen]

Things really go pearshaped only below 100 Hz.

So 257Hz is borderline OK from a distortion point of view. Also the C2-79 is slightly milder with distortion in the 100Hz-200Hz band, for some odd reason.

Anyway, I'm still going to insert these smaller caps, to have some more margin to play in, and to deal with the low-end roll-off of the C2-90.

Also a future project (Avalon opus copy comparing a tweaked passive crossover to the THEL active crossover: a benchmark project) requires a higher cross-over frequency.

 

[Jussi]

So 257Hz is borderline OK from a distortion point of view. Also the C2-79 is slightly milder with distortion in the 100Hz-200Hz band, for some odd reason.

Don't know about C2-79 but at least C2-77 is pretty distorted without that highpass. Reaches about 2% THD at 200Hz with 95dB SPL.

Anyway, I'm still going to insert these smaller caps, to have some more margin to play in, and to deal with the low-end roll-off of the C2-90.

Where do you cross C2-90 to tweeter? C2-24 is it?

Jussi

 

[fgroen]

Don't know about C2-79 but at least C2-77 is pretty distorted without that highpass. Reaches about 2% THD at 200Hz with 95dB SPL.



Where do you cross C2-90 to tweeter? C2-24 is it?

Jussi

C2-79 is about the same, 1,5% at 200 Hz.

I cross the tweeter pretty high, around 4000Hz. Just before K2 starts to rise to a 1,5% peak at 5500 Hz.

Have to play with that too. At the moment there's nothing shocking happening in that area so I don't really have an urge to play with that setting right now. At some stage I'll have to get systematic with it and execute a series of settings and determine what the impact is on the sonic performance.

Tweeter is the C2-12. Has to be crossed above 2000Hz (can push it to 1300-1500 Hz with some more distortion).

 

[fgroen]

Just checked out the C2-24. Is also available with C2-12 front plate. Interesting!!! More efficiency than C2-12 and also very smooth.

At the time I bought the componentry for this project, the C2-24 was not yet available (I think... ) Anyway, it hasn't been considered...

Since I still have to get a pair of C2-12's for the Avalon Opus Clone project, I strongly consider getting a pair of those C2-24 suckers and using the C2-12's for the Opus.

 

[Robert GS]

Very nice loudspeaker - but

Hi fgroen.

I think it's a very nice set of loudspeakers you've build

Sorry but I have to say this:
You should measure your speakers otherwise you don't know if you have highest possible fidelity these speakers can give - and always go for an active solution and no bi-amping or tri-amping with passive filters in between.

Regards
Robert

 

[fgroen]

Measuring will definately be done in the near future.

They are already active. No bi- or tri-amping nonsense in my listening room. That's like buying the gear and not using it's potential.

 

[fgroen]

I'm just checking out the link in your foot note. Cool stuff!!!!

 

[Robert GS]

Thanks, I think so too - and more important it sounds nice

 

[Jussi]

C2-79 is about the same, 1,5% at 200 Hz.

Is there measurements from it somewhere? Accutons own are with 90dB SPL or so and at least responses are smoothed.

Since I still have to get a pair of C2-12's for the Avalon Opus Clone project, I strongly consider getting a pair of those C2-24 suckers and using the C2-12's for the Opus. [/B]

Yep. More efficient, neo motor.. Praised units but I don't have any reliable first hand information from them. But propably I'm just poor and yellowbelly.

Jussi

 

[fgroen]

The accuton website is not accurate (although in the bottom of the pages on individual drivers are measurements performed by Hobby Hifi's Bernd Timmermans)

Also the Accuton website is desperately obsolete and doesn't conatin any info on recent driver model.

Happily our Dutch importer Clofis is more actual and provides good info.

Although in Dutch, just click on the link corresponding to the desired driver type.

 

[mrtech]

active sys

Hello fgroen
i saw a lot of avalon based projects -yours looks
the best.
i understand that you are in the "side" of active crossover
and used it in the avalon .
i am thinking about active system.
1.do you have any exp. or information about using
the hypex ucd moduls in active speakers?
2.did you measure the 12680 eton driver
if yes how close was the fs to the produser spec?
thanks
moshe

 

[fgroen]

Re: active sys

Hello fgroen
i saw a lot of avalon based projects -yours looks
the best.
i understand that you are in the "side" of active crossover
and used it in the avalon .
i am thinking about active system.
1.do you have any exp. or information about using
the hypex ucd moduls in active speakers?
2.did you measure the 12680 eton driver
if yes how close was the fs to the produser spec?
thanks
moshe

Sorry for late reply. Was away for a couple of weeks.

Thanks for your appreciation.

Regarding UcD's: if you're enthousiastic about these amps, they will also work in an active set-up. Make sure all amps are identical. "Dedicated" amps for lows, mids and highs will not give an homogeneous sound.

I didn't measure any driver, and just relied on the manufacturer's specs.