Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Multi-Way Conventional loudspeakers with crossovers

celestion 66 needs Mid-range
celestion 66 needs Mid-range
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Thread Tools Search this Thread
Old 16th July 2013, 02:58 PM   #971
alan-1-b is offline alan-1-b  United Kingdom
diyAudio Member
Join Date: Apr 2006
Location: Midlands
Default winding Inductors , and other about Inductors ; Electrolytic caps ; Turntable motors

'ullo Nigel ,

no , there is no difficulty "in building and inductor" if one understands the particular application the inductor is to be used in ,
that is , R.F. tuning versus Power Supply Regulation and Filtering versus Passive Loudspeaker Crossovers , all require a somewhat different allocation of priorities.

I have been posting about choices in buying ready made here , because that is what Reggie prefers to do , and perhaps DennyG ? ,
though LucasAdamson was originally intending to wind his own , but now ?

If one understands the inevitable trade-offs when using finite materials in combinations for a complex mechanical + magnetic + electrical application ,
and if one has the manual dexterity to control the winding , together with the patience to endure the time it takes to wind many many turns ,
then one can make one's own , as I do.
It would be useful to have 3 hands , and for my next attempt to wind I intend to engage the assistance of another person so that a suitable adhesive can be applied to the wire whilst I am winding ,
because I want a coil that will have no internal localized vibration , and there-by has no modulated magnetic field other than that caused by the music signal.

As I have been posting in this thread , for Series connected inductors in woofer filters one needs low DCR with respect to the DCR of the woofer , ideally a 1:10 ratio or higher ,
though for a 4 ohm woofer as Celestion's crossing in the low midrange , the physical size of the coil plus the cost of the wire renders that ratio impractical for many DIYers ,
so we are making do with a 1:5 ratio that includes the Total DCR of both the Series connected coils.
Ideally this Total DCR should include that of the loudspeaker cable also , thus the users here will likely have reduced to 1:4 or lower ...
and really I consider 1:4 including the cable's loop DCR to be the minimum for reasonably well controlled bass.

Inductors in Parallel arms of midrange and tweeter filters do not need very low DCR , and often are more useful in the circuit if have some DCR of up to 20% of the DCR of the driver
IF the driver's peak-to-peak voice-coil excursion is sufficient to tolerate a small reduction in the filtering of low frequencies at the maximum sound output level required of the driver.

For your 600 Hz application you will need 16AWG or thicker wire for a woofer filter , depending on the inductance required ,
and 20AWG or thinner wire for the Parallel connected inductor in the 600 Hz Highpass filter filter.

I think Active filtering is better for crossing over at low midrange and mid-bass frequencies , and that includes up to at least 400 Hz ,
but also to your 600 Hz depending on the drivers , available amplifier quality , and your budget.
My preference is to cross in the upper bass , between 120Hz <---> 160Hz , to a midrange driver that will tolerate playing that low ,
and for that I will be designing an Active crossover -{ if I stop spending my available time posting in Forums ! }.

*** *** ***

Electrolytic capacitors:

as with types of Inductors , their application to quality audio depends on the type of circuit they are to be used in.
In a circuit where the caps are included in a entirely Electrical Feedback loop , electrolytic caps can be got to deliver closer to transparent audio
than in circuits that include mechanical to magnetic to electrical Back-EMF , which is what passive crossovers have to cope with.
Nigel's reported favourable results with electro caps are in the circuit types they can be got to work well in ,
AND , where a poor quality polypropylene cap , such as one that is some-what microphonic , would cause the sound to be much worse.

*** *** ***

Nigel ,

as I read in your Profile on this web-site that you are a Turntable designer , I think it likely you will know more than a little about motors ...
I have two NOS samples of a Papst motor of the type that was apparently used in the Mark II version of the Oracle turntable years ago.
It is a DC Servo type , and one which can be got to work across a wide range of speeds.
I have forgotten its Model Number , but I will look later , and post it in a suitable thread in a suitable Forum of diyaudio.

I will need to design for a speed that will provide sufficient Torque via the pulley to belt to sub-platter diameter to start and maintain steady speed of a likely heavy platter.
I prefer to use a heavy platter so as to achieve high rotational Inertia , and there-by as low as possible slowing under load whilst an LP is playing ,
because I cannot stand listening to any wow&flutter , as that not only wobbles the Pitch of notes ,
but also it reduces the impact of Transients , that is it causes dynamics' related compression of a sort albeit via the effect on the stylus' cantilever.

As you will know , in any mechanical system there is a fundamental Resonant frequency of vibration.
My concern here is that the motor assembley itself will have a resonant frequency which will be excited when the motor is operated at a speed very close to that frequency ,
and that resonance will then interfere with the stability of the motor speed , and particually so with the Servo which will be then constantly causing changes to the induced speed errors ,
thus I will have to find that Resonant frequency and not use the motor at that speed.

The resonant frequency of the motor assembley will change to some degree when the motor is bolted to the turntable plinth , depending on the tightness of the coupling ,
thus there will be a second Resonance generated if the motor is operated at a speed near to that frequency ,
hence I am going to have to find a speed I can operate the motor at which does not excite two or more resonant frequencies , and which provides sufficient Torque ...
all before I can decide on a pulley diameter for a Gear Ratio to a subplatter.
Now , if you have grappled with this predicament , please can you post the Titles and Authors of any Threads in diyaudio or elsewhere where this is discussed ?
... and similarly for Papst DC Servo motors ?

Yes , motors cause vibration regardless of the speeds they are operated at.
I can find some ways to minimize the effects of some vibration.
What I want to do is avoid speeds which coincide with resonant frequencies in the related mechanical structures ,
because those will cause greater amounts of vibration than other speeds.

Lower speed is only lower vibration when no resonance is excited ,
and lower speed does not necessarily provide sufficient Torque ...
assuming I have understood the Torque requirements for audio turntables , though I may not have ..?

Last edited by alan-1-b; 16th July 2013 at 03:10 PM. Reason: to change one phrase
  Reply With Quote
Old 16th July 2013, 03:59 PM   #972
nigel pearson is offline nigel pearson  United Kingdom
diyAudio Member
Join Date: Mar 2007
Location: Oxfordshire
Hi Alan . If it is allowed a discussion about motors . There is one great fly in the ointment and also saving grace of belt drives . They reduce the transmission of vibration . The idea was to make very cheap transcription turntables when AR did the first commersial design . Very simple tests will show that in a typical belt drive only power output of the motor matters . The belt is almost like an osmotic barrier . The only problem is the vibration of the stylus gets absorbed . One must be honest and say the trade off is excellent . There comes a point when the trade off is the limiting factor .

Stiffening the belt is the answer . The leader tape of old works well .

The remote location of the motor seems a reasonable thing to do . Also twin motors or a tentioner ( like motorcycle valve chain adjuster ) .

Platter mass might be ideal at 3 kg . I do not place much faith in the mass route . Having said that some sound fine . Verdier seems an excellent design . I know all the maths etc . Still am not convinced . Bakelite is a great material . Acrylic also . Bronze outer perhaps ?

Tom Fletcher I think found the motor to be the least important component as long as the power output is high and vibration low . A speeded up direct drive via a stiff belt seems ideal ( as yours ? ) . Garrard used 1500 rpm ( ish ) = 25 Hz . Linn of old 250 rpm = 4.2 Hz . I have a feeling 25 Hz is more OK although can be heard . It is away from other problems like Fo of pick up ( one hopes ) .

My observation is the power output of most motors is too low by a factor of 10 ( LP12 of old ) . One that contradicted that was a turntable using a tape recorder capstan motor . It had an internal speed regulator . My guess is it could mimic a larger motor torque wise . The internal servo being very nice in operation . As the guy said it is the only place where a servo might work . FG servo if memory is correct .

I am 90% convinced the resonance of the motor is a smaller factor than having a belt . I am 90% convinced high platter mass is risky . Dancing with divers boots I call it ( why it matters I don't know ) . I suspect the problem is bearing related . If magnetic levitation is used it should be OK . Still begs the question about doing it .

Optimum use of mass is critical .

I recently bought a JVC direct drive for my girl friend . It is the piece of junk it would be . However it sounds decent . Then I looked again . Very clever engineering . The pick up arm although minimal has made that a virtue in the mass . The motor is powerful and cog free ( within it's concept that is , reduced cog effects ) . The horrible plastic is just wall paper . The chassis is substantial . If it has a defect it masks detail . It is almost good enough for broadcast . I will miss it as it is easy to use . L-E3 . Would love the MC pick up it could have .

The old Lenco 75 was annoyingly good .
  Reply With Quote
Old 16th July 2013, 04:17 PM   #973
nigel pearson is offline nigel pearson  United Kingdom
diyAudio Member
Join Date: Mar 2007
Location: Oxfordshire
The platter mass issue is interesting . Many turntables that have it are as interesting as a funeral . Now that could be because they are doing nothing wrong .

Cutting lathes sound great for playback . They sound nothing like high mass turntables . However they do sound like the master tape ( exceedingly ) .

I had a bronze platter made for a Garrard 301 . An exact replica . It worked fine . It just sounded pedestrian . The bearing is uniquely able to allow this on a 301 . Later much bronze was removed and special plastics used to fill the portholes machined away .That worked nicely ( like Thorens reference designs , thanks Kurt ) .
  Reply With Quote
Old 18th July 2013, 05:24 AM   #974
reggie is offline reggie  Australia
diyAudio Member
Join Date: Feb 2010
Location: melbourne australia
Hi Alan. A quick question please. In your post #969 under your heading Jantzen Audio Inductors at para 2 you say "my primary recommendation for the 3.5mh coil is the 3.4mh coil #35.." But then, in the last sentence of that para you say "..I strongly recommend the more expensive coil #34.." As I am a very literal person (as in I find ambiguity troubling) could you confirm for me please that it is a 3.4mh coil #35 I should be purchasing. Then, if I may, could I post a list of the items I intend to purchase before I actually do it?
  Reply With Quote
Old 19th July 2013, 07:17 PM   #975
alan-1-b is offline alan-1-b  United Kingdom
diyAudio Member
Join Date: Apr 2006
Location: Midlands
Default a Correction to my #969 on Page 97 , about a Jantzen inductor

Originally Posted by alan-1-b View Post

Jantzen Audio inductors:

My primary recommendation for the 3.5mH is 3.4mH: Coil No.35 , 14AWG , 0.488 ohm.
3.4mH will work here because with lower DCR in the circuit there does not need to be as high inductance for the same filtering ,
however , Reggie , I think dropping to 3.3mH is likely to be too low , thus I do not recommend that option.
My secondary recommendation is: 3.5mH , Coil No.1493 , 15AWG , 0.690 ohm ,
however as that is only about half the DCR of Celestion's inductor I strongly recommend the more expensive Coil No.34
so as to get as low as possible DCR in this circuit position where there is no benefit to having any DCR.
In my Post #969 on Page 97 is a mistake.
It is in the section I posted again as a Quote in the above box where I stated:
"however as that is only about half the DCR of Celestion's inductor I strongly recommend the more expensive Coil No.34 ... "
I do NOT recommend Coil No.34 , I recommend Coil No.35 ,
and for the reasons I stated in #969.
Despite proof-reading my post I failed to see the error , perhaps because too much 3.4mH thinking in my brain I typed #34 instead of #35.

Coil No.34 is the 3.3mH which I advised Reggie to not buy.
It would be a good buy for owners of Celestion 44 , as that model had a 3.3mH inductor in the woofer filter.
I apologize for whatever confusion this mistake may have caused to readers.
As I have on several occasions during this thread later posted Corrections to earlier posts I advise here that readers look always before doing anything
to see if I later posted a Correction , particually after my long posts because there it is easy for me to miss a typo as my brain is focussing on the subject more than on my spelling ...
and this latter is a weakness of mine also.

Thankyou Reggie for drawing my attention to the mistake ,
and hey , after me catching you with a schematic mistake a while ago , now you have caught me !
Well , this type of thing is common , and a substantially more experienced designer of integrated circuits , etc , than I am of loudspeaker systems ,
made the point in his book about the use of having one's collegues proof-read one's work before it is published ,
and that included for them to check his circuit designs at work , and each others' also.

Yes , do post a complete Parts List here , and if you are prefering a once only Order ,
then consider including the 24uF or 25uF caps for the midrange filter in either SoniCap or ClarityCap
in case your Mundorf 22uF cap filters off a little too much of the lower midrange - that being a central part of the Male vocals' range.
Cost of later mailing two caps from overseas may be excessive versus the possible no extra mailing cost if bought now with other parts.

*** *** ***

'ullo Nigel ,

Thankyou indeed for all that you posted about Turntable design - I am grateful !
I will post a bit of follow-up to that when I next return to here.
I have little time available today - I posted primarily today to correct my earlier Inductor part number mistake.
  Reply With Quote
Old 20th July 2013, 03:02 PM   #976
nigel pearson is offline nigel pearson  United Kingdom
diyAudio Member
Join Date: Mar 2007
Location: Oxfordshire
Seeing as someone might know . Is there a simple way to modify a standard drive unit for open baffle use ?The Celestion 44 might make an interesting design if having a spare bit of MDF . I have no idea of Qts of a 44 , might be 0.7 . About 1.3 is what the OB people choose

Here is one I fancy using if it will adapt . No problem building a current drive amplifier if that is what it needs .

LS00430 - PRO SIGNAL - WOOFER, POLYPROP CONE, 8OHM, 8 | Farnell United Kingdom
  Reply With Quote
Old 20th July 2013, 10:21 PM   #977
moermusic is offline moermusic  Canada
diyAudio Member
Join Date: Feb 2009
Location: Alberta
bump back to the 66's. Just found a pair in Vietnam for $2000 PLUS $111,115 shipping, all in Canadian cash !!!!!!!! seller has turned down 3 offers, but who could ever afford them in North America. now, back to you.
  Reply With Quote
Old 25th July 2013, 08:21 AM   #978
reggie is offline reggie  Australia
diyAudio Member
Join Date: Feb 2010
Location: melbourne australia
Hi (once again) Alan.
Iíve made my list of parts (and some extra questions).
Caps: Utilise existing 2 x 68uf caps with additional 1 x 8.2uf cap from tweeter
Coils: Jantzen Air-Cored - 1 x 3.4mh, 14awg No.35 and
1 x 2.2mh, 14awg No.28
Resistors: Mills MRA5 - 1 x 1ohm, 1 x 1.5ohm, 1 x 8.2ohm(i)

Caps: Jantzen Superior Z caps - 1 x 3.9uf, 1 x 22uf(ii)
Coils: Jantzen Air-Cored Ė 1 x 2.2mh 14awg No.28(iii) and
1 x 0.35mh 18awg 0.27ohm(iv)
Resistors: Mills MRA5 - 1 x 1.8ohm, 1 x 3.9ohm

Caps Jantzen Superior Z caps - 2 x 1.8uf(v) and
2 x 5.6uf(vi)
Coils: Jantzen Air-Cored Ė 1 x 0.15mh 20awg 0.23ohm(vii)
Resistors: Mills MRA5 Ė 1 x 1.2ohm, 1 x 15ohm(viii)
(i) If, in the future I change the 8.2uf capacitor, will this resistor also change?
(ii) Note that you have suggested I may need to use a 24 or 25uf capacitor here. (Although I liked the change to the 22uf from the 30 odd uf I had in there before.)
(iii) Should this coil be the same awg as the coil in the woofer. Jantzen do make 2.2mh coils in other wire gauges?
(iv) Again, we havenít discussed replacement of these mid-range inductors. Is this an appropriate coil?
(v) Cross-over calls for 1 x 3.6uf but Jantzen doesnít have one. Their closest is 3.3uf or 3.9uf.
(vi) Again, Jantzenís closest is 10uf
(vii) Alan, at your post #902 you recommended a 0.15mh, 20awg coil for the tweeter. Which is good because Jantzen donít make a 0.15mh. (actually, I have taken your post out of text). Is this size coil ok with these capacitors?
(viii) Your recommendation for the LPad was for either a 1.1ohm and a 15ohm resistor, or a 1.2ohm and an 18ohm. Should I order both formats?

I know this is going to cost me (relatively) big dollars but, will be the last time, surely??
Thank you Alan
  Reply With Quote
Old 25th July 2013, 09:43 AM   #979
nigel pearson is offline nigel pearson  United Kingdom
diyAudio Member
Join Date: Mar 2007
Location: Oxfordshire
Ready made chokes 10 mH . What's my best price ?
  Reply With Quote
Old 26th July 2013, 09:52 PM   #980
alan-1-b is offline alan-1-b  United Kingdom
diyAudio Member
Join Date: Apr 2006
Location: Midlands
Default Replies to posts since my #975

'ullo Nigel ,

the short answer is I recommend you do not buy that LS00430 woofer from Farnell , for several reasons ,
and there is better for what you may be designing , for little higher in price.
I will reply more about this below.

*** *** ***

Hi Doug ,

well , you know one person who could afford 66s in North America -
- the guy you sold your new caps' 66s to for too low a price ...
and now realize what he may on-sell them for if he sees that Vietnam advertisment !

What loudspeakers are you listening through now ?

*** *** ***

G'day Reggie ,

Your "Woofer" list is correct , but as for the other drivers !?!

To reply to your questions:

(i) - No

(ii) - the 22uF reduced the excess lower mids , however after you upgrade the woofer filter and note that the woofer had been reproducing some excess of the lower mids also ,
you may hear that the lower mids are a little too much reduced with 22uF.
Ideally I think you should listen with the 22uF Mundorf in circuit after the woofer and tweeter circuit upgrades , and decide then ,
especially as you may be wasting money on that expensive Jantzen 22uF cap.
If you are determined to buy Jantzen caps , then buy three of 8.2uF and connect them in Parallel to sum to 24.6uF ,
but really that is both a waste of space and money when a 24uF SoniCap or 25uF ClarityCap will do the job just as well for less money and in less space ,
and with less trouble than having to neatly solder a 3 caps' bundle into the circuit.

(iii) - No ! - this coil , as in Parallel , needs some DCR to reduce Resonance in the filter.
I have posted several times about this , especially for Bandpass filters , which a midrange filter in a 3-way crossover is.
There is nothing wrong with using Celestion's inductor there.
Do your samples of those seem to have any faults , such as loose windings ?

(iv) - there is no need to replace this coil , as it seems low enough in DCR.
Let us wait till we see what DennyG may find when he does an LTSpice simulation of the actual Impedance of the MD-500 at 5kHz ,
because a higher Inductance there may cause a better result , particually with the tweeter upgrade you have that adds more output.
Celestion seem to have balanced the upper mids' output in a way to complement the lower output of their tweeter.

(v) and (vi) - you are the fellow who wants to buy Jantzen caps !
I have been recommending ClarityCap and SoniCap - what is it about those you do not favour ?
Also , connecting caps in Parallel pairs in a circuit for treble requires one to solder very neatly and with the pairs of cap leads in such way as to not cause Series Inductance
or other losses ... if one really wants to be fussy ... and that would be the only sensible reason to want those large size and expensive caps.

(vii) - Celestion's tweeter inductor is OK , unless your samples have loose windings .?.
There is only one other reason to change it - that is if one does not want to buy 3.6uF and 11uF or 12uF caps.
If you want to buy 3.3uF and 10uF , then change the inductor to 0.15mH ,
but post here about that so I can find one that is not too low in DCR , because for a Parallel circuit position cap some DCR is beneficial.
The one I saw on the Speakerbug list is slightly too low , and Jantzen do have a few that seem closer to optimum ,
thus as you are having a Jantzen inductor special ordered for the woofer filter you may as well for the tweeter filter also ,
if you are determined to change that inductor so as to use 3.3uF and 10uF caps.

(viii) you may have read my #902 again , but it seems you may not have read my #937 and #938 again ,
as I calculated the L-Pad resistor values for an Impedance that used the 0.14mH Celestion inductor in circuit.
If you want to use 0.15mH I will re-calculate the L-Pad resistor values for the different circuit Impedance.
Please post which option ?

"I know this is going to cost me (relatively) big dollars but, will be the last time, surely??"

The Last Time was a song by the Rolling Stones , which I think you may know ,
but as for when will be "the last time" for High-Roller Reggie ... ?

Before I speculate more about your crossover components , please inform me ,
do you intend to make a new board for the woofer filter only , and use the old Celestion board for the mid and tweeter filters ,
or do you intend to make one new board for the entire crossover ?
I ask the latter part , because if you are planning to put sensitive inductors for mids and treble
on the same board as the new large Magnetic Field radiating woofer inductors you will need a much larger board ...
... which may not fit through the woofer cut-out into the cabinet ... though you could make two new boards ...

*** *** ***

'ullo again Nigel ,

'oi , geezers like wot we are being Engineers of a sort wind our own Inductors ,
and really we do not ask about prices without specifying a target DCR , or the particular location and application in the Circuit ...
I'll take a guess as you want 10mH that it is for Series to a woofer ..?
and thus if you want to not lose signal and ruin Damping you will need low DCR wrt the woofer Impedance ...
Well , to get about 10% of a typical woofer's DCR in 10mH you will likely need a 12AWG wire inductor if air-core.
That may be about 5 inches or larger in Diameter , depending on its Height ,
and may weigh about 4 Kilo or more ...
thus maybe you will prefer to chance what can be achieved with Jantzen's P-core inductors ... which will not saturate as badly as Ferrite cored ,
but whether close enough to performance of an Air-core to suit your hearing I do not know ,
however if you are intending to use that cheap Farnell 8 inch woofer it will not matter , because you won't get much bass from it in an open baffle.
You will need to use at least two per channel if 8" to even hear much bass in a Small room , and even then it will be fairly low volume.
Two 10" per channel would be better , but one 10" may suffice if you intend only an experimental project and don't expect a lot of bass SPL.
Before I post more , what is you room size and the result you hope to achieve ?

Last edited by alan-1-b; 26th July 2013 at 10:05 PM. Reason: to change some words and add a few others
  Reply With Quote


celestion 66 needs Mid-rangeHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

New To Site? Need Help?

All times are GMT. The time now is 05:38 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.79%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio