Bottlehead Forum
General Category => General Discussion => Topic started by: Paully on January 19, 2011, 11:24:35 AM
-
Grainger and I replaced the crossovers in my Altecs last weekend. It was a truly dramatic change for the better, much more so than I even thought it could be. Damping the horns didn't hurt either. But of course there were no markings on the inductors. I would like to find out how close they were to the stated value of 3.5mh on my schematic. I can't be certain, from what I am told and have read, that the schematic really reflects what was in there and I would like to know. So if I pay shipping there and back, is anyone willing to go to the trouble who has the tools to measure one or two? Thanks.
Paul
-
Paul,
could you let us know what you ended up doing? I have a pair of 846As that may someday rebuild. It is encouraging that you improved them!
Thanks
-
First I pulled the horns out and damped them with Dynamat xtreme. It made a difference and was worth the money. I think I paid $50 for 4 sheets but only used 2 that were 3 square feet each (plenty left over to damp my turntable). Then we pulled out the crossovers. I used a Russian PIO for the 4 mfd, didn't particularly care what I used for the 10.5. I replaced the 3.5mh inductors with ones from PartsExpress. On the advice of Zilch at the Lansing Heritage site I used an air gapped one for the high frequencies and an iron core for the low. Instead of replacing the pot and resistor with better ones, also on the advice of Zilch, I used a 16 ohm L-pad to replace them both. Anyway, to say I was pleased with the results would be an understatement. And it was nice to have Grainger there who also seemed to be fairly stunned at how well things turned out as well. I didn't think there was as much room for improvement as what I got. Two concerns, I have only listened a few hours since, so you never know what I may come to feel in a month or so. Second, some concern as to what the real value of the stock inductors is. Also, keep in mind that some have said that you should replace the crossover with a model 19 crossover. No idea, just mention it as something you could read up on.
Oh yea, I also replaced the fiberglass insulation with some acoustic eggshell stuff also from PartsExpress. What difference that made I couldn't say because we did it at the same time as the crossovers.
-
Hi Paul,
It looks like my 846A will get a second chance. I think I'll need to buy an extra house to use them though!
Would it be much to ask about the inductors and L-pad part numbers?
And also, did you do something to brace the cabinets or improve the sealing?
Thanks again,
Xavier
-
I can't remember if the 846A was ever made in an 8 ohm version or if that was just the 846B but I am sure you will know what the impedance of your speakers is for the L-Pad. Anyway, whether or not these are the best choices these are what I used! Getting the L-Pad to fit was a chore. We ended up drilling into a new spot in the wood with a fastener bit to wallow out an indent on the outside of the cabinet where the nut could sink down onto the shaft and then drilled out the rest of the hole and mounted the L-Pad on the inside. Hard to describe.
http://www.parts-express.com/pe/showdetl.cfm?Partnumber=255-278
http://www.parts-express.com/pe/showdetl.cfm?Partnumber=266-916
http://www.parts-express.com/pe/showdetl.cfm?Partnumber=260-254
http://www.parts-express.com/pe/showdetl.cfm?Partnumber=260-516
Didn't do any bracing or sealing. I know that is recommended. Maybe a project for another day. The Acoustic Foam stuff I bought worked really well in terms of dimensions. Easy to cut and the remnants I was able to use on the back of the speaker cabinet. If you get around to this, do let me know how it works out for you and what you think. It is an involved project. Once I get them from Grainger I can forward on detailed photos of the crossovers to look at.
-
A quick Google search will show a bunch of hits on measuring inductors. The link below is one of them. I don't have a decade resistance box and I presume you don't have one either. Since you are trying to confirm the 3.5 mh value, I would simply reverse the formula to calculate the value of the resistor and then see if that value resistor works in the circuit. At least it would be a starting point.
Of course, since the formula is dependent on frequency, this would be the inductance at 60 hz.
http://www.coilgun.info/theoryinductors/measureinductance.htm (http://www.coilgun.info/theoryinductors/measureinductance.htm)
ray
-
Just a thought, but couldn't one use a potentiometer in place of the Resistor box to get the equal voltage measurement then measure the resistance across the pot for use in the formula?
-
Paully is looking for someone who owns a device that will measure inductors who will measure it for him and ship it back to him.
On the Coilgun suggestion, I think Paully is right. A pot wired to one outer leg and the wiper will suffice for a decade box. You just have to measure it after getting a null rather than read it from a decade box. The pot must not be inductive or you are in trouble.
But what really has me scratching my head is that this looks like it will only measure the series equivalent resistance not the reactive (inductive) component of the "Unknown" inductor.
This is somewhat of a "Bridge" to measure impedance (meaning, complex, AC and reactive). I don't see how this method measures the reactive part of the unknown.
And maybe I'm wrong. Maybe PJ will take a look?
-
Grainger,
The WT3 Woofer Tester will do it. I probably have another instrutment to do it - somewhere. You probably have my contact info. You can send it to me if you like.
So - how is that Seduction power supply?
-
Paul,
I have woofertester that El Greco just mentioned which reminded me that it can measure RLC values. I'm in Colorado and will be glad to ddo it for you, but there may be somebody closer.
Thanks for the notes on your speaker rebuild -- sounds very much like what I plan to do for my Cornwalls, but also with the bracing and sealing -- may even flush mount the drivers if it looks possible. Everything about these speaks is original and the only things that need work are the tweeter horns (which barely make 10 KHz now, the caps in the crossovers, and the seals on the midhorns. That will be a summer project though. Cabinet and grille are immaculate -- showroom condition.
-- Jim
-
I used to use something similar to this "Coil Gun" set up to measure the load on Constant Voltage Speaker lines.
IIRC, it used a 400 Hz Square Wave as a test signal, rather than the 6.3 VAC from a transformer, but the principle was the same.
And, yes, Paully, I expect a pot would do in place of the decade box, but I think you would want linear taper, and you may need more than one value to zero in on the null point. IOW, keep reducing the pot impedance 'til you get the smallest value that will get you to the null point, since the smaller value will give you a more accurate reading across its range.
-
I may not understand, but I have two inductance meters (one only inductance, and a multimeter with an inductance scale) that I have been using to measure chokes and crossover inductors for years, they sure seem to measure exactly or close to the values marked on each inductor or choke. Is this not an accurate or proper way to measure chokes and inductors???
Cordially,
Lee Hankins
-
Thanks for the parts links Paul.
And please do send me the crossover pics. I have a long list of projects but This one definitely made into that list!
-
lth1, I would suspect the answer lies in how your meters arrive at their results. But, the short answer is, IMO, if your meters are reading close to the known value of a part, they must be doing something right! At least, within the range of values you are interested in.
-
Man I just realized - I am a 'newbie' still after all these years (thanks Doc, I'm just kidding)
Hope all you's oldie's and er, well newbie's are doing well.
Time for a road trip Pauly and Grainger? It's only a 3 hour drive.
-
Jim, thank you for the reply. The way Paully started out this Post and the responses he was getting I just thought that I was missing something or doing something wrong. Most newer DMM's seem to include an inductance range so why would one consider sending out the inductors, freight would cost more than an inductance meter.
Lee Hankins
-
Well, I can't say that I have much experience with the inductance-measuring feature on newer meters; I certainly don't own one with that feature, but I can sure see how it would come in handy.
Seeing these responses and links, to me, reinforces the idea that measuring inductance is considerably more complex than, say, measuring resistance or Voltage. In my mind, anyway, the meter would have to be generating an AC test signal, measuring some aspect of how the inductor reacts to the frequency it is generating, then translating that into a useful reading on the display! Quite a bit of math going on there, among other things!
-
Man, I had a beautiful and very long-winded answer, which failed to post and I lost it!
Oh well. The fact is, an inductor with an iron core will have an impedance that is partly inductive, partly resistive, and partly non-linear (hysteresis losses). And these will vary with frequency and signal level. (And at high frequencies, capacitances will mask inductive behavior making measurements impossible.
For crossover applications, the best practical thing is to measure the series resistance, the inductance, and the parallel "resistance" at a frequency near the crossover frequency and at a signal level near that expected in use. For this you need a bridge that works with an external oscillator, and of course the external oscillator. These buggers are difficult to use. I don't have one.
-
elcraigo,
The Monster power supply got stalled. Trouble in the high voltage supply side. That and cutting and changing the feed to the C4S board for dual mono feed. I hate to take my Seduction out of the system to butcher it while it sounds so good.
I will do it soon, I keep saying that but will.
-
I am really, really happy with the sound of the speakers from what I have heard to this point. Ecstatic actually. But I don't need to be able to measure inductors multiple times (though I appreciate the suggestion, it was neat). I just want to know what it is for these one time since I probably am not going to become a crossover engineer (anyone who is familiar with my posts knows that isn't going to be a serious problem for me anytime in the near future). Postage back and forth is pretty much all I want to incur. So Jim, if you will send your address to me and you don't mind sending them back at my expense, I would love to know how close the inductors are that I took out to the schematic that is on the web that everyone seems to be using. I know there is a lot more to it, but I still would like to see if they measure close to the stated 3.5mh. I appreciate the offer, thank you!
And Jim, you may not be interested, it sounds like you know plenty about what you have planned for your speakers, but when I upgraded my Klipsch crossovers for my Chorus II using Bob Crites' I was pretty pleased.
-
Paul, congrats on the rebuild. I'm in the process of building new crossovers for my Model 19's. I've got a handheld LCR meter and was astounded by how much my inductors were off by, a couple as much as 25%! The caps were even worse at up to 50%. No wonder I've needed a balance control on my preamp... I went all out and picked up North Creek inductors and all of the caps and resistors are within 1% matching.
I'm waiting on speaker wire to build the crossovers and plan on burning them in with my Paramounts once the soft start kits come out before putting them into the speaker. My to do list includes Dynamat for the horns, internal bracing and resealing the cabs. Hopefully my results are as good as yours!
-
Paul,
Sent you a PM the other day. Haven't heard from you so don't know if you saw it.
-- Jim
-
Paul may not have turned on email notification for PMs. I sent him an email.
-
"I've got a handheld LCR meter and was astounded by how much my inductors were off by, a couple as much as 25%!"
I am really surprised by that, one, because all of the parts in my speakers were still within 10% and also I wouldn't have thought that an inductor would drift all that much. I guess I think of them as little transformers where the winding is what matters and that shouldn't age so much. But I also don't know that much about inductors! Just curious what would cause an inductor to drift over time or is it possible that, just like the 846A, they changed the crossover values here and there with the Model 19 while tweaking and didn't publish it. That was why I was so curious to measure the value of my inductors. I have read that this specific crossover for the 846A is often found with widely varying values that have nothing to do with drift, so I wanted to know if mine matched up to the schematic or not.
-
Depends on the Inductor
I've seen inductor specs from -30 /+50% to +/- 3%
-
Inductors don't drift, but their measured value (if they have a non-air core) will vary with frequency and measuring current. And with exactly what is measured, since the terminal impedance is a combination of several things with different phase angles - some measurements sort these out differently from other measurements. The long-winded answer I wrote, which got lost (see my previous post) went into excruciating detail - probably a good thing it was lost!
Bottom line, never trust an inductance measurement unless you know an awful lot about it and the device being measured! I have a simple-minded choke checker which I use for plate and power supply chokes, and it regularly reads 10% to 30% lower than proper, carefully made measurements done with good instruments. Since I expect that error and have confirmed that it is always with that range of error, it's still useful to me - but it's sure not accurate!
-
As PJ said, I don't have the right tool ;)
I read somewhere that the tolerances with the Altecs was lower than 10%. For some reason 25% comes to mind but I'm probably wrong on that too....
-
Paul,
YGPM, and your inductors will be on the way back to you today.
-- Jim
-
As PJ said, I don't have the right tool ;)
I read somewhere that the tolerances with the Altecs was lower than 10%. For some reason 25% comes to mind but I'm probably wrong on that too....
I got the results from Paul. They are low by about 25%. I think you got it right.
-
Grainger,
Yes, they are low, but at least they are quite close to each other. I ran the test many times on each and got very consistent results.
I had been thinking of selling this little gadget and the software, but no way I'm doing that now -- this thing is just too cool. I'm going to have to ressurect mycritical headphone cable project.
-- Jim
-
Glad to know I'm not completely crazy ;)
Finished one crossover last weekend and am working on the second now. Attached is a pic for size reference. Just a little difference is all :)
-
Building a crossover with high quality components takes a lot of space. Great picture. Is that about a 4:1 ratio.
-
No joke! I had thought of using the same layout as the original but the heat sinking factor of the large gauge wire made it difficult to solder. I picked up a hydraulic crimper from harbor freight and put 8 tons of force on the crimps instead :)
-
What wire did you decide to use?
-
If that's directed at me, I'm using Mogami 12g OFC speaker wire from Redco.com. They have a great selection of wire for decent prices (no affiliation of course). I've been using the Mogami wire as speaker wire (and interconnect) for years now and am happy with the sound of it so thought it natural to put it all the way through the system.