‘Tale of a Tweeter/Equalizer Dilemma’

[frankmarsi]

2-1-07

Dear CSP members, hello and I’m still recovering from the ‘house-fire’ that occurred here a few days ago, re; the 94 replies. I dare say I had written a response to the over-all point just after K.K.’s response, but at the last minute decided not to post it as I sensed I’d only be adding fuel to that fire. Besides maybe everyone was just experiencing a ‘senior-moment’, or sitting on a tack? Honestly, I felt everyone was entitled to their respective opinions. I really wouldn’t want to see anyone leave this celebrated site because of a misunderstanding or difference of opinion.

But here I am again in with another question.

When I bought over $800. worth of AB-T__h high range tweeters for my 4- LST’s and 4-3a’s and then appealed to everyone here about those tweeters’ excess of mid frequencies and its refusal to actually match the originals frequency response as-they radiate far too much mid-frequency information for a tweeter. Some members suggested using, damn the thought; an ‘equalizer’, some also came forth with resistor and capacitor circuits. Well I did try using different capacitor values dumbly enough directly across the tweeter leads, 4.7mfd having a decent effect, although some very high frequencies were seemingly ‘shaved-off’.

Fast forward, thru months of sitting by in quiet desperation looking for a plan to get me out of my tweeter dilemma I created while justifying buying all those damn tweeters out of fear they would be discontinued. My alternative might have been ‘Dyna-Audio’ or ‘Scan-Speak’ or some other ‘pricey’ replacement, which my budget won’t accept.

So, I picked up a S.A.E.1800 parametric equalizer, which I believe may help, although I feel if I’m trimming mid-frequencies to my new tweeters, I might also be cutting off some information to the high frequency top-end as well and also killing the info to the working mid-range speakers too. Compound all that craziness by the fact that I have a Revox A77 and a Pioneer CTF900 cassette deck (I can’t forget all those precious cassettes I made in the late ‘70s on TDK- SAX tape), the problem being the SAE unit only permits using one tape deck at a time, as this only has one set of inputs? Any suggestions on using two decks, could a ‘Y’ plug help?

While we’re here, might there be any suggestions about another means of curtailing the excess mid-range information from these damn tweeters without touching the X-over?

Stumped and idle in AR-ville.

Frank Marsi

frankmarsi@verizon.net

[Carlspeak]

Frank:

My guess is you have 3 options listed below in order of increasing complexity;

1) Sell the tweeters back to AB-Tech

2) Use the equalizer - much to your dismay

3) Modify the xover by raising the xover point or, adding a parallel trap circuit. This type of circuit is intended to mitigate 'humps' in the frequency response. It is composed of a resistor, capacitor and inductor, all wired in parallel with each other and as a group, in series with the tweeter.

It sounds from your descsription like you have a response hump based on your description of too much mid range from the tweeter. This, of course, assumes you are well within the normal pass band of the tweeter's operation. Are you certain the tweeter xover point isn't set too low because of the new tweeter's incompatibility with the factory xover?

Did you get this problem with the AB Tweeters running off your existing circuit? If so, perhaps you need to raise the xover point a bit to suit the new tweeters. Or, as a last resort, get someone to help you with the design of the parallel trap circuit.

In any case, you need to have one of your speakers frequency response tested with the new tweeters installed in order to quantify exactly where the problem lies in the frequency spectrum. With that information you can then decide whether to raise the xover point or, design the parallel trap circuit. There is no easy fix for this one.

It's all about the music

Carl

Carl's Custom Loudspeakers

[johnieo]

Well I did try using different capacitor values

>dumbly enough directly across the tweeter leads, 4.7mfd having

>a decent effect, although some very high frequencies were

>seemingly ‘shaved-off’.

Frank: If you placed any capacitance in parallel with the tweeter it would undesirably affect the tweeter's high frequency performance as you observed.

The problem many folk observe with the AB-Tech tweeter is that it extends too low into the midrange. To increase the tweeter crossover (reduce its low mid output) one would change from an RC crossover to an L-C design. The LC crossover L(parallel) C(series) with tweeter is used in the AR-11, AR-3a-Improved, and AR-3a Limited, with varying component values in each design). Adding capacitance directly across the tweeter leads will not achieve your goal.

It is a pain, but no substitute for removing woofer and getting out a soldering iron!

[onplane]

>2-1-07

>Dear CSP members,

>

>Well I did try using different capacitor values

>dumbly enough directly across the tweeter leads, 4.7mfd having

>a decent effect, although some very high frequencies were

>seemingly ‘shaved-off’.

>

>So, I picked up a S.A.E.1800 parametric equalizer, which I

>believe may help, although I feel if I’m trimming

>mid-frequencies to my new tweeters, I might also be cutting

>off some information to the high frequency top-end as well and

>also killing the info to the working mid-range speakers too.

>Frank Marsi

>frankmarsi@verizon.net

>

Hi, Frank!

Like John said, a cap across the tweeter will seriously reduce the tweeter's high frequency response.

Frank, have you tried replacing the 6mfd cap in the tweeter xover with one of those 4.7mfd caps?? This should lower the current going through the tweeter network. Specifically, it will add approx. 1.3 ohms of resistance at 5000 Hz.

Now, the 4.7 cap will reduce current in the tweeter network for all frequencies, but the more dramatic reduction will occur in the lower frequencies.

Regards,

Jerry

[frankmarsi]

Carl, it seems more and complicated every answer I get, but thanks.

Let me finish crying first and then hopefull I have a different mood.

Thanks, Frank Marsi

[frankmarsi]

Go-on John, you apparently seem more than correct about my tweeter problem. I don't want to cause any resentment here as I've mentioned this problem of mine before, but you're answer is painfully correct, I'm just looking for and easy way out. Unfortuately there is no easy answer. I must enter the 'big-bad-box' of my AR-LSTs' and AR-3a's.

Thanks, Frank Marsi

[frankmarsi]

Jerry, Jerry, you're so..... right! As I stated above I was only trying for a 'quick-fix'- saddenly it's not gonna happen for me is it Jerry and everyone else whose replied to my problem? It just pains me so cause this set-back might be a set back even longer as I may have to relocate in a few weeks and everything will have to be packed up and moved to a new location. I fear of the move possibly be a little un-settling for all o=f my equipment. The wood cabinets and the electronics, heck this stuff is old. Good thing is, I will have two rooms of substancially larger size and both of my systems will thank me for that.

Thanks, Frank Marsi

[dynaco_dan]

Hi Frank;

I remember when you posted that you had bought those particular tweeters.

My suggestion is because time has passed since you bought them, allow more feedback ideas before you grab a screwdriver and go into the depths.

As was suggested earlier, seeing if A-B Tech would buy them back at minimal loss to you.

Selling them privately to someone local with no shipping and handling may be an option.

[soundminded]

Frank, since everyone else has given up, let me stick my two cents in just for the heck of it and because I've been experimenting with tweeters for awhile now. If your system has a separate preamp you are in luck. I'd do something like what I did experimenting with the 901s. Don't put any capacitors in parallel with the tweeters, this will just roll off the high end. Use the preamp output to drive a separate integrated amp connected ONLY to the tweeters. Use the equalizer in the tape monitor circuit of the integrated amp. Only a blocking capacitor is necessary for the tweeters but if it has a second order crossover associated with it that's ok too. Of course be certain the integrated amp is connected ONLY to the tweeters, any connection between the two amps and you risk blowing one or both up. If the intergrated amp has a bass control, turn it to minimum. Use the integrated amp's level control, its treble control, and the equalizer to adjust the level and frequency contour of the tweeters. Also try reversing their polarity. Don't get discouraged if you don't get good results at first or if you like what you hear at first and then later change your mind. IMO, there is an instinctive tendency to initially adjust the overall balance of a system to be too bright. The equalizer should make it possible for you to filter out the excess midrange output from the new tweeters without affecting the midrange drivers. Good Luck. BTW, with 901s it took me 2 1/2 years so be patient. Rome wasn't built in a day.

Also don't get too shaken up by our recent 94 entry fire storm, it was a tempest in a teapot. My philosophical disagreements with Ken go back a lot longer. Here is a thread that's over a year old;

http://www.classicspeakerpages.net/dc/dcbo...mesg_id=7&page=

I'm sure there's a lot more to be said about this topic and I don't expect we will ever agree with each other. That's OK, it would be a pretty dull world if everyone thought the same way. I will defend Ken Kantor to the death in his right to be wrong :-)

[onplane]

>Frank, since everyone else has given up, let me stick my two

>cents in just for the heck of it and because I've been

>experimenting with tweeters for awhile now. If your system

>has a separate preamp you are in luck. I'd do something like

>what I did experimenting with the 901s. Don't put any

>capacitors in parallel with the tweeters, this will just roll

>off the high end. Use the preamp output to drive a separate

>integrated amp connected ONLY to the tweeters. Use the

>equalizer in the tape monitor circuit of the integrated amp.

>Only a blocking capacitor is necessary for the tweeters but if

>it has a second order crossover associated with it that's ok

>too. Of course be certain the integrated amp is connected

>ONLY to the tweeters, any connection between the two amps and

>you risk blowing one or both up. If the intergrated amp has a

>bass control, turn it to minimum. Use the integrated amp's

>level control, its treble control, and the equalizer to adjust

>the level and frequency contour of the tweeters. Also try

>reversing their polarity. Don't get discouraged if you don't

>get good results at first or if you like what you hear at

>first and then later change your mind. IMO, there is an

>instinctive tendency to initially adjust the overall balance

>of a system to be too bright. The equalizer should make it

>possible for you to filter out the excess midrange output from

>the new tweeters without affecting the midrange drivers. Good

>Luck. BTW, with 901s it took me 2 1/2 years so be patient.

>Rome wasn't built in a day.

>

Soundminded, passive bi-amping is a possible solution, but these LST's are something else. The autotransformer in the xover network makes any kind of passive bi-amping a little more difficult.

It can be done, but will require some thought on how to separate the network so that the load left on the autotransformer doesn't foul up the remaining drivers.

Frankly, I think if you go down the bi-amp trail on the LST's you should go the whole way. That is, I'd eliminate the autotransformer, the switches, and the huge cap bank. Then I'd have a separate amps for the woofer (1), mid-drivers (4) and tweeters (4).

In all honesty, I get a headache just thinking about these LST's. I believe soundminded is on the right track, but it's NOT going to be easy and I don't think you can go "half way".

Regards,

Jerry

[dynaco_dan]

Hi again Frank;

I would never give up on you but it has been a painful and lengthly and complex workaround offered to you, either method.

If I didn't know the complexity of the AR-LST crossover and how you would have to basicly gut them to install those new tweeters.

You would also need to spend much more money with more separate electronics.

In most cases this would leave a non-reverting system if you should decide to return to the, "old way", and buy used older drivers, later.

Be darned if you do, be darned if you don't.

Re-sale value of them after all of these mods should also carry some weight in your decision, Frank.

Ditto with the AR-3A's.

I already wrote my story here and to you directly regarding my facing the replacement of all 8 LST tweeter drivers 10 years ago and the end result.

I was presented with a few options at that time.

[soundminded]

Jerry, I'm not sure I understand what you mean by "passive bi-amping." I forgot about the transformer. In that case, I'd disconnect the tweeters from the transformer altogether and just use a simple capacitor to block the low frequencies from damaging them. Are they 4 ohm tweeters? Is a series parallel combination of 4 of them appropriate to keep the load at 4 ohms? (do they all play at the same volume?) Here's a handy dandy calculator for figuring out what capacitor for a first order high pass or coil and cap values for a 2nd order high pass would be.

http://www.the12volt.com/caraudio/crosscalc.asp#cc

Once the system is adjusted to Frank's satisfaction, the master volume control on the main preamp will control everything, the other controls can be used for occasional tweaking.

Here's one word of advice, DO NOT plug the integrated amp into a switched outlet on your main amp or preamp. Its inrush draw current each time you turn it on will eventually damage the power switch on your main preamp. (I did this once already.) Use a separate power strip.

BTW, I wired all of my tweeters in parallel and used a separate capacitor for each one. On some of my setupts it's under 1 ohm but the tweeters draw so little current it doesn't seem to matter. In my case, I used 8 ohm tweeters. I've got three of them in parallel per channel through a 3.3 mfd each, all in parallel with my AR9s. I think about 5 in parallel with each of the empires. In both cases this is in addition to the main tweeter. In both cases, they are across the same amp as the main speaker but I've never burned out an amp...at least not that way.

[Carlspeak]

Frank. Here's another thought on your tweeter dilemma. It's not a solution - just an observation.

You would think that if someone drops $800 on an order for tweeters for their LST's that AB Tech would be sufficiently familiar with their product to be aware of the problem they will cause and provide a solution UP FRONT. After all, they are the claimed AR experts.

I worked in quality engineering as a CQE for much of my working career and learned early on that getting to root cause of a problem is the best route to take as opposed to treating a symptom - which is what many of the posts here are addressing. Actually, the best root cause solution would be to sell LST tweeters that are a better match with the originals and won't cause the problems you describe.

There are some pretty savvy AR guys who post here. Some, I suspect, may be more knowledgeable than the guys at AB Tech. Perhaps they could lean on Carlos or Jason a bit to help their customers a little better by providing comparative specs. on the drivers they sell vs the drivers they are intended to replace. Or, in your case, at least provide a schematic of how to mod the circuit to correctly raise the crossover point for the new tweeters.

Blind faith sometimes leads you down a dark alley. Buyer beware!

It's all about the music

Carl

Carl's Custom Loudspeakers

[kkantor]

Actually, reversing the polarity of the tweeters is a good idea... worth a try.

-k

kkantor.spaces.live.com

As one Master Baiter said to another:

"I will defend Ken Kantor to the death in his right to be wrong."

"There must be some way to arrange this."

[kkantor]

If somebody wants to loan me one of those tweeters for a couple of weeks, I'll try and develop a quick and dirty mod to it. (Probably some kind of series capacitor.)

Meanwhile the suggestion to try reversing the polarity of the tweets is a good one.

Yeah, an EQ will probably be of some help and the LST takes EQ better than most speakers, due to its wide dispersion. But it will be hard to get the EQ to sound just right, especially with a garden-variety graphic equalizer.

-k

kkantor.spaces.live.com

PS- on a personal note, sorry the theatrical thread was disturbing to you, honestly. To me, it is all part of life in a virtual community. If it is any solace, the topic had the highest views/age ratio of anything I can remember on this site. I'd wager the net effect on subscriptions was neutral-to-positive. Mark can correct me if my guess is wrong.

[dynaco_dan]

Hi Ken;

I am expecting Mark to now say, "now that membership and readership has increased, thank's to that firey debate, we need to have more of them".

Thank you, from me, for coming back to leave your pleasant thoughts and also clear the air, Ken.

[soundminded]

>If somebody wants to loan me one of those tweeters for a

>couple of weeks, I'll try and develop a quick and dirty mod to

>it. (Probably some kind of series capacitor.)

>

>Meanwhile the suggestion to try reversing the polarity of the

>tweets is a good one.

>

>Yeah, an EQ will probably be of some help and the LST takes EQ

>better than most speakers, due to its wide dispersion. But it

>will be hard to get the EQ to sound just right, especially

>with a garden-variety graphic equalizer.

>

>-k

>

>kkantor.spaces.live.com

>

>

>PS- on a personal note, sorry the theatrical thread was

>disturbing to you, honestly. To me, it is all part of life in

>a virtual community. If it is any solace, the topic had the

>highest views/age ratio of anything I can remember on this

>site. I'd wager the net effect on subscriptions was

>neutral-to-positive. Mark can correct me if my guess is

>wrong.

>

>

Same here. No hard feelings on my part whatsoever. (Do you presume you are the first person in cyberspace I've had this argument with? I'm well practiced at it. I know what you are going to say in response to my postings long before you actually say it. :-) ) Consider this possibility Ken, it's not that I don't know what you are saying or don't appreciate it in fact I once believed most of the things you say, it's that I've reconsidered their importance in light of a larger context.

Reversing the polarity of the tweeter offers some slim hope of slight improvemnt due to cancellation in the crossover region where the peak might be. The problem the transformer presents is that based on my understanding of it from the literature in the library, it adjusts the woofer and tweeter simultaneously. Unless the replacement tweeters are an exact match to he original in every respect including sensitivity, impedence, FR, even dispersion, the results will be different and likely detrimental, certainly unsatisfactory compared to the original which has been optimized. Biamplifying with a separate equalizer may be difficult but at least it offers the possibility of adjusting the tweeter output independently of the rest of the system and once a match is found, the settings can be left that way. Correction with a passive network using the existing transformer may require some resistors for output level matching and an LC network for shaping FR as well. I'm sure you're up to that Ken, I know you can do it.

[Carlspeak]

>If somebody wants to loan me one of those tweeters for a

>couple of weeks, I'll try and develop a quick and dirty mod to

>it. (Probably some kind of series capacitor.)

>

>Meanwhile the suggestion to try reversing the polarity of the

>tweets is a good one.

>

>Yeah, an EQ will probably be of some help and the LST takes EQ

>better than most speakers, due to its wide dispersion. But it

>will be hard to get the EQ to sound just right, especially

>with a garden-variety graphic equalizer.

>

>-k

>

>kkantor.spaces.live.com

>

>

>PS- on a personal note, sorry the theatrical thread was

>disturbing to you, honestly. To me, it is all part of life in

>a virtual community. If it is any solace, the topic had the

>highest views/age ratio of anything I can remember on this

>site. I'd wager the net effect on subscriptions was

>neutral-to-positive. Mark can correct me if my guess is

>wrong.

>

There were indeed a large number of viewes. Soap operas do have a way of attracting viewers.

It's all about the music

Carl

Carl's Custom Loudspeakers

[soundminded]

>There were indeed a large number of viewes. Soap operas do

>have a way of attracting viewers.

>

>It's all about the music

>

>Carl

>Carl's Custom Loudspeakers

Other than the gawk factor, everyone loving a good fight, it just might be remotely possible that there was interest in the topic itself.

Should I feel like it's David against Goliath, lil' ole me the underdog going up against an MIT grad-U-ate? Don't forget, David won. :-)

[administrator]

There are a few things consistent about threads like the one I locked:

1.) I become aware of them because I get emailed.

2.) I get complaints if I kill them.

3.) I get complaints if I don't kill them.

That's why I'm very interested in implementing a user based moderation system, kind of like Slashdot - where users can easily attribute a score to each post and then each user can set their "filter" accordingly.

If anyone has strong feelings about which forum software they would like me to move to, you can send me an email at webmaster@arsenal.net. A pre-requisite is obviously the user based moderation functionality.

Mark

[soundminded]

I'd like to make it clear here and now that however my exchanges with Ken Kantor may have been perceived, they were intended as good natured chiding if somewhat rough and tumble and not to be mean spirited. That's kind of the culture I came from in New York City so please take that into account. I also want to make it clear that I have nothing but the greatest respect for him both personally and professionally and celebrate his achievements during his distinguished career. I am also very greatful that he posts here. There is nobody I'd rather disagree and argue with...no matter how pigheaded he sometimes gets. :-) Besides, our discussion isn't ended by a longshot. There's still much to explore and I'd hate to lose the opportunity were he to leave.

My message to Mark had nothing to do with him, it had to do with someone else impersonating me by stealing what appeared to me to be my moniker. If I am going to be seen to insult someone, I want it to be the genuine article.

[Carlspeak]

>There are a few things consistent about threads like the one

>I locked:

>

>1.) I become aware of them because I get emailed.

>

>2.) I get complaints if I kill them.

>

>3.) I get complaints if I don't kill them.

>

>That's why I'm very interested in implementing a user based

>moderation system, kind of like Slashdot - where users can

>easily attribute a score to each post and then each user can

>set their "filter" accordingly.

>

>If anyone has strong feelings about which forum software they

>would like me to move to, you can send me an email at

>webmaster@arsenal.net. A pre-requisite is obviously the user

>based moderation functionality.

>

>Mark

FYI all viewers,

Here is a link to the Slashdot FAQ web page where moderation is discussed in some detail. After wading thru this you should have some understanding of what Mark is proposing.

http://slashdot.org/faq/com-mod.shtml#cm600

It's all about the music

Carl

Carl's Custom Loudspeakers

[administrator]

vBulletin seems to have the features I want, is currently being developed, and I should be able to convert all the content from this forum software into it. It's not quite Slash, but may be a good compromise between what I run here (DCForum+) and Slash.

Here's an example of the software:

http://www.vbulletin.com/forum/

Mark

[roundhome]

>If somebody wants to loan me one of those tweeters for a

>couple of weeks, I'll try and develop a quick and dirty mod to

>it. (Probably some kind of series capacitor.)

>

>Meanwhile the suggestion to try reversing the polarity of the

>tweets is a good one.

>

>Yeah, an EQ will probably be of some help and the LST takes EQ

>better than most speakers, due to its wide dispersion. But it

>will be hard to get the EQ to sound just right, especially

>with a garden-variety graphic equalizer.

>

>-k

>

>kkantor.spaces.live.com

>

>

>PS- on a personal note, sorry the theatrical thread was

>disturbing to you, honestly. To me, it is all part of life in

>a virtual community. If it is any solace, the topic had the

>highest views/age ratio of anything I can remember on this

>site. I'd wager the net effect on subscriptions was

>neutral-to-positive. Mark can correct me if my guess is

>wrong.

>

>

would you like me to send my LST crossover to you to test also I have it out right now?