AR-10pi Bass Compensation, Voicing, Design Details

[soundminded]

To clarify, my end tables are McIntosh ML1Cs, not AR 10 pis.

[Guest Bret]

>I'd prefer to get back to the technical issues if your still interested<

Pete,

I didn't think it would be very kind of me to just "disappear" from the thread as you might have taken that as a rejection.

Truth is, Pete, I'm just tired of discussing it at all. There have been a lot of glittering generalities thrown around, but the real, true, bottom-line is that anyone who tries to do a restoration job on these old speakers is taking a shot in the dark no matter what they do.

Take for instance the tweeters: What are the exact measurements of the old tweeter? We don't know. What are the measurements of the replacement tweeters? We don't know. Has the glue dried and affected the performance of the old? We don't know. How can we test? We can't. If they are different, how can we compensate? We can't.

The capacitors: Is a Bennic a reasonable replacement for a Sprague Compulytic? We don't know. Why not? We have no numbers on the Sprague. Well, we should stay with electrolytics shouldn't we? All things being equal, yes. Are all things equal? We don't know. Can we find out? No. Is a Solens the wrong way to go? Maybe. Maybe Zen's are better? Without double-blind A/B tests we can't tell. Right. Even then we'd only have a consensus "best," but would that be right? We don't know. Is there any way at all to find out? No. Would it benefit some future others who might face the same choices were we able to arrive at a consensus as to the most effective substitution? Perhaps.

The woofers: Do we know what the operational parameters were on the original woofer for this cabinet? Pretty much; we think; maybe, depending on which version of the same-numbered woofer was in there. Do we know which version, exactly? No. Was there a change while the units were in production? Yes. Did the woofer remain within the specs we have that we can find? We don't know. How stiff was the suspension? Stiffer. Stiffer? Is that a little stiffer, or a lot stiffer, or what? We don't really know. If we did know, would it matter; could we rebuild the drivers to exactly match the goal? No. Why not? Because we don't know the stiffness of the spider, we have no original surround, there are no OEM replacement parts available so we will change some parameter when we repair the speaker. How will that effect the sound? We don't know.

I originally thought that having this discussion in public might be beneficial to the "tribal mind" as well as me, personally. It seems to have done little but spark sometimes caustic debate.

I appreciate your willingness to go back to square one and begin again, but I don't have the will or energy to spare in a "silly" pursuit of a restored Chevy.

And yes, I'm afraid that a lot of professional musicians, conductors, and others would say that in their day, these ARs were the Jaguars. Well, okay, maybe not. . . but I won't go any further down the list than Volvos.

I'm thinking that they wouldn't have had to be mass-produced if they weren't any good. The fact that a pair of 10pi's cost the equivalent of say $3,000-4,000 in today's dollars and yet had to be mass-produced I think speaks rather highly of them.

Until next time, Pete. And thanks for all the research you did to discover some of the differences in the drivers and so-on. I still say it's great that anyone would be that interested in the plight of another.

Bret

[Pete B]

Hi Bret,

I'm about ready for a break myself sometimes it helps to take a fresh look after a break. I should get back to some real work anyway.

>Truth is, Pete, I'm just tired of discussing it at all. There

>have been a lot of glittering generalities thrown around, but

>the real, true, bottom-line is that anyone who tries to do a

>restoration job on these old speakers is taking a shot in the

>dark no matter what they do.

I think that with some guidance these issues can be resolved. I think we understand the woofer situation much better. And perhaps we're going to have to accept a wider tolerance since most restorers do not have a lab to recheck the measured parameters, or perhaps trust our ears more once a few sanity checks are done.

>Take for instance the tweeters: What are the exact

>measurements of the old tweeter? We don't know. What are the

>measurements of the replacement tweeters? We don't know. Has

>the glue dried and affected the performance of the old? We

>don't know. How can we test? We can't. If they are

>different, how can we compensate? We can't.

Each issue can be worked out one at a time. There are the very early frequency response plots for the AR-3a that could be used as a reference. It would also help to measure a decent size sample of the older tweeters. But as I understand it the 10pi uses the newer tweeter and we know the Fs of 2 kHz, that's a start.

>The capacitors: Is a Bennic a reasonable replacement for a

>Sprague Compulytic? We don't know. Why not? We have no

>numbers on the Sprague. Well, we should stay with

>electrolytics shouldn't we? All things being equal, yes. Are

>all things equal? We don't know. Can we find out? No. Is

>a Solens the wrong way to go? Maybe. Maybe Zen's are better?

> Without double-blind A/B tests we can't tell. Right. Even

>then we'd only have a consensus "best," but would that be

>right? We don't know. Is there any way at all to find out?

>No. Would it benefit some future others who might face the

>same choices were we able to arrive at a consensus as to the

>most effective substitution? Perhaps.

I've been wanting to write more about capacitors but I'll have to find the time. ESR among a type of cap is not that different, electro vs. electro etc. We should ask are the newer polys important and consider that AR never used a poly IIRC, so why all the concern? I don't believe NHT went any better than mylars, that should be a hint. I would use paralleled Bennics where the Compulytics were used, I can give you an educated estimate if your interested. Interesting that you mention blind A/B cap tests you might be very surprised with the results if this group attempted it.

>The woofers: Do we know what the operational parameters were

>on the original woofer for this cabinet? Pretty much; we

>think; maybe, depending on which version of the same-numbered

>woofer was in there. Do we know which version, exactly? No.

>Was there a change while the units were in production? Yes.

>Did the woofer remain within the specs we have that we can

>find? We don't know. How stiff was the suspension? Stiffer.

> Stiffer? Is that a little stiffer, or a lot stiffer, or

>what? We don't really know. If we did know, would it matter;

>could we rebuild the drivers to exactly match the goal? No.

>Why not? Because we don't know the stiffness of the spider,

>we have no original surround, there are no OEM replacement

>parts available so we will change some parameter when we

>repair the speaker. How will that effect the sound? We don't

>know.

Suspension stiffness and Fs are not that critical in these smaller AR-3a sized boxes as I've stated before. Mass and motor strength are much more critical and these requirements can be determined based on box volume and published Fc, and motor strength from published system Qtc. It's interesting that suspension stiffness *is* more critical with the 9s because of the larger box and therefore lower alpha. Unibox predicts the difference in sound. I believe we figured most of this out in the other thread, or at least a start. A larger sample size would help. Let me offer an educated guess about motor strength, the higher strength probably made up for the losses in the complex AR-9 XO, and the autotransformer in the 10pi.

>I originally thought that having this discussion in public

>might be beneficial to the "tribal mind" as well as me,

>personally. It seems to have done little but spark sometimes

>caustic debate.

I'm surprised by the hostile reaction, I think we made a lot of progress in the woofer thread and also the midrange thread.

>I appreciate your willingness to go back to square one and

>begin again, but I don't have the will or energy to spare in a

>"silly" pursuit of a restored Chevy.

I didn't like the "silly" comment either and I don't think this is arguing just adjusting perspectives a bit. I believe that what is unfortunate is that there's such a hostile reaction when people don't follow the tribal "group think" which I think is held by a fairly small but vocal number here.

OK, well perhaps I pushed too far to the other side in trying to make a point, how about Mercedes as an analogy but not the Gullwing? Seems like it should be an American brand given AR's heritage. I will give you that the 3a, 10pi, LST, 11, and 9 were up there in the product line but they're still mass produced. I also realize that they're much better than the 2ax I mentioned, still this example does show that there was not a high priority given to on axis response.

>And yes, I'm afraid that a lot of professional musicians,

>conductors, and others would say that in their day, these ARs

>were the Jaguars. Well, okay, maybe not. . . but I won't go

>any further down the list than Volvos.

>

>I'm thinking that they wouldn't have had to be mass-produced

>if they weren't any good. The fact that a pair of 10pi's cost

>the equivalent of say $3,000-4,000 in today's dollars and yet

>had to be mass-produced I think speaks rather highly of them.

Shows that production costs have come down, interesting comparison.

>Until next time, Pete. And thanks for all the research you

>did to discover some of the differences in the drivers and

>so-on. I still say it's great that anyone would be that

>interested in the plight of another.

Enjoyed the challenge, I think we're getting there. It's good to see that you and a few others are not just blindly following the "formula" for replacement parts but are doing something about the differences in the sound.

Yes, until next time,

Pete B.