Pundit
Post by harvey on Jul 14, 2022 17:44:31 GMT 12
Next up is a new centre channel and fronts for my HT system, I purchased most of the drivers last year then put the project on the backburner for a while. I plan on building a set these: Zaph Audio ZDT3.5Most of the crossover parts are under control but the design specifies a 3.6Mh 16 AWG Erse Super Q inductor, I’ve just found out that Erse have gone out of business and they are unobtanium. I have found a lesser quality replacement but they will likely run close to $100 landed and I do have a spare pair of 3.0mH 16 AWG Erse inductors that I replaced out of my main 2ch speakers. Is there anyone out there that can wind some more wire onto these to bring them up to the required spec for me?
|
Post by RdM on Jul 14, 2022 20:18:13 GMT 12
Next up is a new centre channel and fronts for my HT system, I purchased most of the drivers last year then put the project on the backburner for a while. I plan on building a set these: Zaph Audio ZDT3.5Most of the crossover parts are under control but the design specifies a 3.6Mh 16 AWG Erse Super Q inductor, I’ve just found out that Erse have gone out of business and they are unobtanium. I have found a lesser quality replacement but they will likely run close to $100 landed and I do have a spare pair of 3.0mH 16 AWG Erse inductors that I replaced out of my main 2ch speakers. Is there anyone out there that can wind some more wire onto these to bring them up to the required spec for me?
Or wind a separate 0.6mH and attach it in parallel series? ( Aargh!;-) Or wind a new 3.6mH air core pair?
Are the Erse air core?
Or wind a new 0.6mH air core yourself?
You can get square bobbins, or you could DIY a circular one? Motor re-winders might be able to supply a small amount of copper wire in whatever gauge, for a small consideration, or buy wherever. The Zaph link implies that larger gauges are even slightly better.
Formulas re wire size, diameter, number of turns, length of wire required, resultant inductance and DCR are in old textbooks, and online:
I imagine one might want to think a bit carefully about placement of the added new 0.6mH coil rel to the existing 3.0mH one...?
Magnetic fields ...
|
Pundit
Post by harvey on Jul 14, 2022 22:04:10 GMT 12
Thanks RDM. The inductors are iron core wired in series with the lower woofer. Required resistance is critical and very low at 0.2 ohms hence the iron core.
I had replaced a couple of inductors in my 2 channel speakers with Goertz foil ones. An expensive exercise that did make an improvement but not enough to justify the money (live and learn), so do also have a couple of spare Solen 14AWG litz wire air coil inductors and was thinking I may be able to use the wire from one of these.
I checked out the formulas you linked to and my head hurts. Seriously I could probably fumble my way though it but would much prefer the peace of mind with having them measured to ensure they are the correct spec but not sure who could do this.
|
Post by RdM on Jul 14, 2022 22:47:02 GMT 12
Thanks RDM. The inductors are iron core wired in series with the lower woofer. Required resistance is critical and very low at 0.2 ohms hence the iron core. I had replaced a couple of inductors in my 2 channel speakers with Goertz foil ones. An expensive exercise that did make an improvement but not enough to justify the money (live and learn), so do also have a couple of spare Solen 14AWG litz wire air coil inductors and was thinking I may be able to use the wire from one of these. I checked out the formulas you linked to and my head hurts. Seriously I could probably fumble my way though it but would much prefer the peace of mind with having them measured to ensure they are the correct spec but not sure who could do this.
OK, existing iron cored. Need only to add just a small extra 0.6mH in series. May as well make air cored, calculate exactly.
You have Solen 14AWG litz wire air coil inductors that you don't know the inductance of? Easy to find out with the right measurement devices, even cheaply available, (I'm too tired to look up right now sorry!) on TM, Ebay etc. Later.
Save all that. Wait... those devices are quite cheap, useful, easy to find.
Let me see if I can find relevant specs formulas for a 0.6mH inductor with whatever wire size...
But don't hold your breath ... it might take a little while.
Having your own measurement device is useful too of course.
But the formulas to say you need so many turns of this gauge around that diameter (air) core to make that inductance are quite clear, solid, you could get close. Then check with the measurement device. Learn how to use it as well of course.
Could it be cheaper to buy one and measure for oneself? Well maybe. Cost of postage, etc. to send coils to check. Check yourself?
Hmm. Just thoughts! ;=})
So I might look up what it takes to wind an 0.6mH inductor, a decent gauge of wire, air core, that you can wire in series, and think further on it, and get back to you.
Others might have even better ideas, contributions ... ?
Warning;- I've been drinking a bottle of wine, so don't trust anything I say;- check it out for yourself! ;=})
|
Pundit
Post by harvey on Jul 15, 2022 9:43:46 GMT 12
Haha I do know the spec of the Litz wire air coil I have and in theory could wind a 0.6ohm coil but DCR of an air coil is just too high (Solen make a 06.ohm 14 AWG air coil and DCR is 0.24ohms). So I am stuck with using an iron core inductor to keep to DCR to the correct value. I understand that some multimetres can measure inductance, either that or you need an oscilliscope and frequency generator and I don't have either. Probably a good time to explain my practical skills far outweigh my technical. I could probably muddle my way through it with the correct equipment but can't justify buying one and faffing around learning how to do it for what is likely to be a one off exersize. Hence my original request, and just to clarify I'd be ok to pay a professional if needed as long as the cost isn't to prohbitive (but just not sure who would be interested in doing this) or pay a suitable hobbyist in liquid refreshment if needed. 
|
Pundit
Post by peter0c on Jul 15, 2022 10:44:58 GMT 12
The aim probably is to have a DCR as low as possible rather than the 'correct' value. This is a problem in some speakers (e.g. Tannoy which has an autoformer) in that the calculations for the other components especially resisters is based around the inductor DCR. (In the same way you need to take into account the driver coil inductance and subtract this from the calculated inductance of a coil in series with the driver for a low pass section). However when looking at your Zalph crossover circuit, the four resistance values given in series with the inductors for each drive unit are the measured resistance values of the inductors, not added resistance. Whilst a resister in series with the tweeters might make some sense (drop the output a bit), it makes little sense to 'throttle' down the woofers. So I suspect the values given are a rare bit of honesty rather than a necessity. You can get a low DCR by either using iron core inductors (cheaper because they have fewer turns but prone to saturation so best avoided) or an air core of 12 AWG (expensive). I'll look into my large stash and see whether I can come up with a 3.6mH iron core of at least 14 AWG and pm you over the weekend.
|
Post by RdM on Jul 15, 2022 13:47:56 GMT 12
Haha I do know the spec of the Litz wire air coil I have and in theory could wind a 0.6ohm coil but DCR of an air coil is just too high (Solen make a 06.ohm 14 AWG air coil and DCR is 0.24ohms). So I am stuck with using an iron core inductor to keep to DCR to the correct value. I understand that some multimetres can measure inductance, either that or you need an oscilliscope and frequency generator and I don't have either. Probably a good time to explain my practical skills far outweigh my technical. I could probably muddle my way through it with the correct equipment but can't justify buying one and faffing around learning how to do it for what is likely to be a one off exersize. Hence my original request, and just to clarify I'd be ok to pay a professional if needed as long as the cost isn't to prohbitive (but just not sure who would be interested in doing this) or pay a suitable hobbyist in liquid refreshment if needed.
Maybe you can contact them and see if they are actually available, & what freight would cost. (I see USPS has some international suspensions, so would have to be a freight company.)
All that said, if you aren't using the 3.0mH you already have for anything I can see the financial attraction of just winding on extra turns to make up to 3.6mH. Where are you? Are there any transformer winders in the area? You could use even thicker 12 or 14 AWG to keep DCR down? Maybe they'd have the ability to test inductance at a low frequency?
I do have a signal generator and scope. I'd have to brush up on theory, technique.
Presumably resonating with a known capacitor. Or aha compare impedance with a resistor. shows how and has handy calculators to make the math easier.
I see the Zaph centre speaker bass unit needs a 1.5mH inductor. More shopping?
There's a graph there of it with different DCR values - not a huge difference? But clearly lower is better.
Cheers!
|
Post by RdM on Jul 15, 2022 16:34:12 GMT 12
First of all, it seems I was wrong about using a hand held LCR type meter for measuring typical bass unit crossovers.
From speakerbug.com.au/index.php?route=product/category&path=18_59
"In his book 'High Performance Loudspeakers', Martin Colloms praises the virtues of Toroids including low distortion and self screening of stray magnetic fields.
Note - The inductance value needs to be measured at 100Hz - The range where they are typically used as an inductor for the bass driver.
Typical handheld LCR meters measure at 1,000 Hz and will read a lower value than labelled."
However that caveat was only on that (expensive!) toroids page. I've also read elsewhere that electronic LCR meters will give false readings on toroids.
In his book Valve Amplifiers, Morgan Jones writes the opposite of measuring air core inductors:
"Many component bridges use a 1 kHz internal oscillator.
When measuring air-cored coils, the inductive component can easily be swamped at low frequencies by the relatively high resistance, causing the bridge to give misleading results. If it is possible to feed such a bridge from an external source of AC, it should be fed with the highest frequency that the bridge manufacturer allows (typically 20 kHz), and this will allow sensible measurements to be taken."
I've got a heavy box of salvaged crossovers on a high shelf ...
I could compare readings on both a large air core and cored inductor and see how the MTester reading compares with my DSE multimeter, which also measures inductance, frequencies unknown. Just for my own intellectual interest, since I hadn't known the above before. It'd be useful to determine their values anyway. I could then compare those readings with those from the sig gen scope and resistor setup. I don't have an actual LCR Bridge as such, but that last does essentially the same thing.
A bit of work, and I've already got a lot on; - I'm not sure that I could tackle your job, but the problem certainly seemed to pique my interest ;-)
Regards!
|
Pundit
Post by harvey on Jul 15, 2022 17:25:18 GMT 12
peter0c I realized resistance in series will affect the sensitivity of the driver but assumed it applied across all drivers in the speaker to get them playing at the same level (hence my comment about the 0.2 ohm value being critical). From the Zaph website “The large Erse inductor was selected for it's low DCR to keep the overall system sensitivity as high as possible” and “the system is sensitive to changes in the series inductor DCR”. I replaced the 3.0mH inductors I now have spare with large (almost the size of a small side plate) Goertz Foil inductor so you could imagine how spendy that was. RdM – Strangely I did order a pair direct from Erse after Parts Express told me they were no longer available, got an email confirming my order including shipping costs but a day later got another a day later saying it had been cancelled. I went back to their website and now they have a minimum order quantity of 10 units. Not sure what is going on there as Parts Express told me they had gone out of business but the Erse website would suggest otherwise. The 1.5mH inductor you mentioned for the centre speaker is in a different position on the woofer circuit to the 3.6mH I am discussing (the centre speaker is a 3 way so doesn’t use the 3.6mH). The tower speaker uses a 1.0mH inductor in the same position, I do have all the other inductors required fortunately. The 3.6mH is wired in series with the lower woofer only and my understanding is this what creates the 0.5 way part of a 3.5 way speaker by preventing that woofer playing some of the higher frequencies and providing a cleaner bass output (happy to be corrected on this). You’re right about the financial aspect of modifying the 3.0mH but it is part only part of the equation. I’d really like to use these now or else I likely never will which would be a shame. Also, there are speaker designers who avoid iron core inductors altogether. Something to do with overloading and saturation points (this is where my head starts to hurt again) but when they are required, for reasons unknown to me the Erse Super Q are the ones recommended by a couple of the speaker designers whom I follow. I’ll have a hunt over the weekend to see if I can find a local transformer winder who may be able to help. Thank you both for the offer to help, much appreciated!
|
Post by RdM on Jul 16, 2022 22:50:10 GMT 12
RdM – Strangely I did order a pair direct from Erse after Parts Express told me they were no longer available, got an email confirming my order including shipping costs but a day later got another a day later saying it had been cancelled. I went back to their website and now they have a minimum order quantity of 10 units. Not sure what is going on there as Parts Express told me they had gone out of business but the Erse website would suggest otherwise.You’re right about the financial aspect of modifying the 3.0mH but it is part only part of the equation. I’d really like to use these now or else I likely never will which would be a shame.Also, there are speaker designers who avoid iron core inductors altogether. Something to do with overloading and saturation points (this is where my head starts to hurt again) but when they are required, for reasons unknown to me Hi, Harvey !
I seem to have messed up the quoting format above; never mind, but so, in order ;- (I had a PC crash & lost already written, I'm reconstructing!)
When was that? The current Erse page for them shows them to be available right now; 1-9 for $15.67 USD ea. (The 12&14g are unavailable, check their links at the end of that page!) _
I only meant that it might be cheaper to add 0.6mH with extra wire. It started me thinking about how I'd join them. And how to guesstimate how much wire, and add maybe 5% just in case.
You could perhaps look at the existing 3.0 mH (5% tolerance, they say - do they mean +/- 5% either side as is common with resistors and capacitors? (resistors can go to much finer tolerances.) ? and count how many turns are visible on the top layer, then - see sideways;= how many layers? Turns on top layer x number of layers. How many turns, approx?
So, with an estimate of total turns (around the core) for 3.0mH, how many more turns approx are needed to add an extra 0.6mH?
Just that percentage more. Add a bit for error. It could be checked later, but you have to scrape the insulation off to measure each check, so good to get an estimate of how many turns first... And try to estimate what wire length is required. Dimensions ...
And think how to joint them. Not just solder side by side. Good metal to metal contact would be ideal.
I'd be tempted to split ends, clean, insert each half, crush together, solder around that. Use prior inserted heat-shrink tubing, &or lacquer/enamel too first?
I'm just guessing!
I thought maybe transformer winders would or might have access to thicker wire? Add the same or thicker (lower AWG#) extra wire for best few 0.6mH extra turns?
They might have know how? But maybe not. Ring around.
But actually that Erse one looks so much cheaper than the equivalent Jantzen one at Parts Express.. _
Yes well saturation. We all know when we've had enough '=}) Coffee, tea, tobacco. Alcohol. Idiot politicians?
Concepts about ferrous metal cores (of several different types) "saturating" - vs air cores which don't - (although they need so many more turns that their DCR goes way up in comparison) -
When a coil is wound around a ferrous metal core, the efficiency goes way up, the inductance per number of turns, and the type of metal affects it too, but still, ultimately, there is a limit ... As the current in the coil increases, the core can hold more, somewhat linear magnetism, but there's an end point where it's fully magnetised, and increased current makes no difference. Ideally if an inductor with an audio signal, or an input or output transformer say for a valve amp, you want the signal to go through without being clipped by hitting saturation of the magnetic properties of that core.
Bigger is better? It can depend. But essentially when the core saturates it means that it can no longer respond to an increasing amplitude signal. It clips, flat-lines, there. Hence distortion.
It's a bit like this that we might have picked up in school days chemistry;- if you stir something soluble in water like sugar or salt, and keep adding it, there will come a point at which no more will dissolve into it.
Then that solution is called saturated. It won't accept any more input. No more will dissolve.
So it happens the same with magnetic cores. So you just need to stay out of that region. Whether the core can take the signal you want to give it.
For your speaker design, they reckon it can. So either see if it's still available, &or see if some extra turns could be reliably joined and wound on? _
These are the re-edited 3rd time thoughts after the PC crashed twice, couldn't edit, had to go to another PC. Aargh!!
Clumsy, I'm sorry! ;=})
Just previous thoughts I'd meant to more elegantly express! I'm sure I could be corrected, but ... I'm just sharing my understandings ;=}) And of course they might be wrong!
_
I found a page on tests of audio inductors core distortions, but it's on the other presently crashed PC in another room. Maybe I should sign up for syncing all these things ...
Meanwhile ... ;=})
~ Ross
|
Post by RdM on Jul 17, 2022 19:04:48 GMT 12
I wrote earlier:
When a coil is wound around a ferrous metal core, the efficiency goes way up, the inductance per number of turns, and the type of metal affects it too, but still, ultimately, there is a limit ...As the current in the coil increases, the core can hold more, somewhat linear magnetism, but there's an end point where it's fully magnetised, and increased current makes no difference.Ideally if an inductor with an audio signal, or an input or output transformer say for a valve amp, you want the signal to go through without being clipped by hitting saturation of the magnetic properties of that core.
Wikipedia explains it better. (see the Effects and uses section too.)
I found a page on tests of audio inductors core distortions
Cheers!
|
Pundit
Post by harvey on Jul 17, 2022 21:33:40 GMT 12
Ok, I’ve gotten to the bottom of the issue with Erse Audio. I can only assume when I saw the minimum order quantity of 10 I was looking at the 3.5mH they have available (I was on my phone), scroll further down and I see the 3.6mH shows available in single units. I found a thread over at DIY Audio with people complaining about having their order cancelled as well with no explanation. Unfortunately the owner passed away and the suggestion is his wife is just running down the inventory. DIY Audio -Erse Thanks for the explanation around saturation, that analogy was great. Anyone that can get me to understand something technical is doing well! So 3.6 is a 20% increase on 3.0. Are we correct in assuming that induction increases in a linear manner in relation to the number of turns? Also, in regards to resistance this relationship does not appear linear. Looking at the Erse website 3.5 mH resistance is 2 ohms and 7mH is 2.9 ohms. I've been working this weekend but have time off tomorrow so will see if we can count the turns successfully and maybe try and find someone who can measure the end result.
|
Post by RdM on Jul 18, 2022 19:32:20 GMT 12
Ok, I’ve gotten to the bottom of the issue with Erse Audio. I can only assume when I saw the minimum order quantity of 10 I was looking at the 3.5mH they have available (I was on my phone), scroll further down and I see the 3.6mH shows available in single units. I found a thread over at DIY Audio with people complaining about having their order cancelled as well with no explanation. Unfortunately the owner passed away and the suggestion is his wife is just running down the inventory. DIY Audio -Erse Thanks for the explanation around saturation, that analogy was great. Anyone that can get me to understand something technical is doing well! So 3.6 is a 20% increase on 3.0. Are we correct in assuming that induction increases in a linear manner in relation to the number of turns? Also, in regards to resistance this relationship does not appear linear. Looking at the Erse website 3.5 mH resistance is 2 ohms and 7mH is 2.9 ohms. I've been working this weekend but have time off tomorrow so will see if we can count the turns successfully and maybe try and find someone who can measure the end result.
Ah ta for the diyaudio link. Sad. The Erse web pages are well organised.
But please give links - do they have the same wire gauges? I briefly browsed...
Resistance - thickness vs length. Lower AWG, larger diameter. Yes, number of turns, but as the layers stack up, the outer diameter of the winding on the core increases, so the wire path length becomes longer for the same one turn...
I was only suggesting a guesstimate figure with peering at side layers and counting the turns on the visible top layer. Use a stylus of some sort, toothpick? to keep count. ;=}))
Then yes an estimate from that ballpark what wire length needed, maybe go over 5-10%. Joining them, may as well source 14 AWG (or even 12 etc.) But 14 AWG more likely.
I sourced some from a jewellery outlet, a small coil, maybe 1m? supposedly silver plated, not sure of length now, unopened. 14 AWG is ~ 1.6277mm dia.
That TM outlet no longer has such thick ones. But maybe its US source does? Soft copper wire for jewellery. But beware of 'German silver'! I hope mine is really silver plated. Else as before other sources of copper wire. Maybe you only need a metre or two. Hence guesstimate, add a bit to be on the safe side, but not too much, etc. Maybe it might have slightly lesser 'purity' than claimed for the original, but you could go the extra 0.6mH with 14 vs 16 AWG as an experiment, might not cost too much, and be reversible.
Anyway, all interesting. I'd bought the TM 14 AWG wire to try a diy 2.4G wifi yagi design, but in the end didn't. I'd made a smaller one with thinner wire, and a 5.8Ghz one with short sections sawn off a telescopic extending antenna, which worked really well.
But that's beside the point;- )
I still haven't checked out my meters on a cored and air core crossover coil yet, but I'll get around to it.
Cheers!
|
Post by RdM on Jul 19, 2022 19:27:34 GMT 12
All that said, if you aren't using the 3.0mH you already have for anything I can see the financial attraction of just winding on extra turns to make up to 3.6mH. Where are you? Are there any transformer winders in the area? You could use even thicker 12 or 14 AWG to keep DCR down?
Again, I don't know where you are, but surely local folk like these would do the same? (North Shore Electric Motor Services)
We have a large range of round and rectangular magnet wire from 0.05mm diameter to 4mm diameter in round wire and numerous rectangular sizes. All our wire has insulation rated to a 180 degrees celsius and higher.
We can de-spool small quantities for low volume requirements.
After getting an idea of how many for 20% more turns may be, then guesstimate the diameter, work out what length. Maybe you'll only need one extra layer?
And if you choose 14 (or even madly 12, but likely too thick, difficult to wind, fewer turns in the width available!?) then its lower DCR for just 0.6mH extra might help? What's the DCR of your candidate previously bought 3.0mH ones, anyway? Maybe just even winding on existing 16 AWG extra brings it to the same 3.6 one spec?
|
Pundit
Post by harvey on Jul 19, 2022 21:48:04 GMT 12
I've got the 14 AWG Solen air core inductors spare that may suit to use, litz wire as well so may be easier to splice on to the Erse inductors.
Sorry for the slow reply, unfortunately I discovered yesterday that the Dayton RS52 Dome midrange drivers used in these speakers have been updated since they were designed, the updated version have issues and may not be suitable to use. Disappointing to say the least but it may mean a change of tack so I'm just going to sort this out first.
|
Post by RdM on Jul 19, 2022 21:52:19 GMT 12
No worries! ;=})
Take your time!
|
