Another AR94Sx 'situation'.

[Guest hilltroll67]

Some time ago I replaced the woofer and mid/woofer surrounds in my pair of AR94Sx speakers, but didn't recap the xovers. I wasn't happy with the resulting sound (bass deep but dull, some sort of problem in the midrange in vocals)and set the speakers aside. Recently I was moved to try the Veselinovic (TNT) xover rebuild project, when a reader of this board inquired about such a rebuild of his speakers.

When I opened up one of the speakers and popped the xover board out, I discovered that it is not the same design as the one Veselinovic worked with. There is no cap in the mid/woofer circuit, and the cap in the woofer circuit bridges the woofer connections. Woofer and mid/woofer each have a 2 Ohm - 22 Watt resistor in series with the plus lead.

I don't really know what all of this means, but if I can reasonably assume that my drivers and the drivers in Veselinovic's speakers have the same specs, I will rebuild one xover with his configuration and the other xover like the one I took out, using Dayton caps in each. And then make a sound comparison.

BTW, the xover I removed uses an iron-core inductor. I will replace that with a Jantzen air core of as close to the same value as I can find (1.40 microHenrys instead of the 1.37 of the original.

Comments, scornful or otherwise, are welcome.

Bob

[Guest hilltroll67]

Well, the sound comparison is highly inconclusive, but the midrange vocals problem is still there. I will now install the Veselinovic style xover and swap out the mid/woofer for a KLH 8" cloth surround driver that probably dates from the '60s. Its cone is actually about 5" dia. compared to the AR's 6", and its resistance across the coil is about an Ohm higher, and its suspension isn't nearly as stiff (even easier to move in and out than the AR5 woofer), but I got the pair of them cheap, and unscientific experimentation is the only kind I can handle.

Bob

[RoyC]

>Well, the sound comparison is highly inconclusive, but the

>midrange vocals problem is still there. I will now install the

>Veselinovic style xover

Hi Bob,

I compeletely agree with your assessment of these speakers.

The "sx" version has different 8" drivers than the AR-94s. I went the same route as you, including trying the 94s crossover, with unsatisfying results.

I ended up using the original crossover with a higher value cap across the woofer, which cleaned things up a bit. I think the main problem is the "midrange" woofer. It is allowed to run full range with no filter whatsoever.

The AR-94sx was definitely not a bright spot in AR history.

Keep us posted...

Roy

[Guest hilltroll67]

Thanks for the info, Roy; it saves me some research. Now I'm hot on the trail of another approach to the problem. Will post the results.

Now I'm getting 'hepped up' about this project. Thanks for getting me started, "res8unck".

Bob

[Guest hilltroll67]

OK. I have done some experimenting. Redid one speaker, we'll call it A, crossover to the 94S/Veselinovic configuration (without the silver wire and electrolytic caps, used all Dayton polypros), and the other speaker B crossover to a conventional design, with an inductor in the mid/woofer circuit, and a different tweeter because of that inductor. I lowered the tweeter crossover point to around 2500Hz, to get a better overlap with the now limited woofer/mid. And tested.

The A speaker has something unpleasant going

on in the upper midrange. I suspect that the untethered woofer/mid driver is clashing with the tweeter somewhere around the 3000Hz tweeter crossover point, but have no hard evidence to support the guess.

The B speaker has no problem in the midrange that I can hear. It's sound level isn't as high as A's in the midrange, probably because the woofer/mid driver is fading at 6dB/octave after around 1800Hz, but the overall balance still seems to be good.

Now that I have eliminated the A setup from consideration, I will change that speaker's crossover to match the one in B, but keep the stock tweeter. Comparing those two setups will (I hope) tell me if AR was unnecessarily cautious in setting it's tweeter crossover point at 3000Hz. If there's no dip in the region between mid and tweeter, there's no reason to go to a different tweeter (like B has).

Sorry if this essay is incomprehensible. It shouldn't be that hard to describe.

Bob

[kkc]

I am only a year and half late replying to this post.....!

I have rebuild the xo on my AR94si speakers. As you noticed, the XO is different from the Dejan/TNT version. I manually traced the XO using a multimeter and visually tracing the wiring. See http://www.own-initiative.com/%7Eowniniti/Misc/AR94_Rebuild.htm for full details.://http://www.own-initiative.com/%7Eow...r full details.

[KlausDK]

I have rebuild the xo on my AR94si speakers.

Hi kkc

I looked through your very nice restauration project, and was puzzled by your statement:

"Reduced the resistance into the woofer to compensate for the higher resistance from the new air-coil"

because in the schematics, the resistance of the new coil is lower than the old (0.63 Ohm -> 0.36 Ohm)?

I don't like resistors in series with a woofer, reduces damping faktor from the amplifier on the driver, have you considered to try without the woofer resistor and then work on the speaker placement to balance the woofer output?

BRgds Klaus

[kkc]

Hi Klaus

I think the resiatance values in the schematic are switched the wrong way. The only way to be sure is for me to open up the speakers and re-measure the air-coil which I will do some time in the future. Thanks for spotting what must be a mistake....!

As for the two resistors, if I remove them the bass will be far too strong regardless of position. Just doubling the resistance (as it was before) reduced the bass so much that the speakers sound pretty bad. Right now the speakers are about 1.5 feet away from the wall and that minimises bass multiplication. I guess without any resistance, I would end up placing the speakers much more into the room which would not work.

Rgds, kkc.

[MikeF]

OK. I have done some experimenting. Redid one speaker, we'll call it A, crossover to the 94S/Veselinovic configuration (without the silver wire and electrolytic caps, used all Dayton polypros), and the other speaker B crossover to a conventional design, with an inductor in the mid/woofer circuit, and a different tweeter because of that inductor. I lowered the tweeter crossover point to around 2500Hz, to get a better overlap with the now limited woofer/mid. And tested.

The A speaker has something unpleasant going

on in the upper midrange. I suspect that the untethered woofer/mid driver is clashing with the tweeter somewhere around the 3000Hz tweeter crossover point, but have no hard evidence to support the guess.

The B speaker has no problem in the midrange that I can hear. It's sound level isn't as high as A's in the midrange, probably because the woofer/mid driver is fading at 6dB/octave after around 1800Hz, but the overall balance still seems to be good.

Now that I have eliminated the A setup from consideration, I will change that speaker's crossover to match the one in B, but keep the stock tweeter. Comparing those two setups will (I hope) tell me if AR was unnecessarily cautious in setting it's tweeter crossover point at 3000Hz. If there's no dip in the region between mid and tweeter, there's no reason to go to a different tweeter (like B

Hey, Bob

What hardware did you end up with on your 94sx rebuild? How did it turn out? Do you have schematics of your final version? I have a pair of 94sx's I bought new and now want to rebuild/refresh them. I could use some specific guidance on the hardware and I'm not finding much relevant and specific information on the 94sx. I'm driving them with a mint Pioneer SX-1010 receiver/amp I also bought new in 1975. If you can get back to me, I will get a shopping list together and get going on the project.

Thanks,

MikeF

[MikeF]

>Well, the sound comparison is highly inconclusive, but the

>midrange vocals problem is still there. I will now install the

>Veselinovic style xover

The "sx" version has different 8" drivers than the AR-94s. I went the same route as you, including trying the 94s crossover, with unsatisfying results.

I ended up using the original crossover with a higher value cap across the woofer, which cleaned things up a bit. I think the main problem is the "midrange" woofer. It is allowed to run full range with no filter whatsoever.

Hi Roy,

Back in time but how did your rebuild turn out on the 94sx's? Do you have a schematic or hardware list to share (more specific, the better)? I'm not finding much help in recent postings and I have a pair of sx's here I don't want to get rid of without at least trying to rejuvenate them.

I appreciate anything that helps me build a shopping list for a rebuild.

Mike

[RoyC]

Hi Mike,

Do you have the original crossovers? I'm going to have to dig around for my notes, but I used the same component values as the original sx crossover, with the exception of one capacitor...which I increased to 75 uf from 40 uf. (It was the cap in the woofer circuit). Since then I've had a good look at an AR-94su, which is different than the "s" or the "sx" versions. It seems AR could not decide what to do with the AR-94. I suspect the earliest version was the best, as it appears later versions like the sx is comprised of cheaper parts and drivers.

These speakers are not worth too much in the way of a restoration effort, imo. It is tempting to part out 94 series cabinets. The tweeters and mid/woofers are useful as replacements in better models.

Roy