Pultec Inductors again...

[maxwall]

Toroid winder link takes me to Goog, is this correct ?

 

[CJ]

It is just a huge toroid being wound, kinda lame.

But here are some A L vs Turns number for a 5 Band Pultec EQ:

A L Calculator

Formula: N = 1000* Sq Root (L/A L) 193.649167310371
(L in milliHenries)

A L Turns 30 mH 50 mH 80 mH 100 mH 150 mH

100 548 707 894 1000 1225
200 387 500 632 707 866
300 316 408 516 577 707
400 274 354 447 500 672
500 245 316 400 447 548
600 224 289 365 408 500
700 207 267 338 378 463
800 194 250 316 354 433
875 185 239 302 338 414
1000 173 224 283 316 387

2000 123 158 200 224 274
3000 100 129 163 183 224
4000 87 112 141 158 194
5000 78 100 127 141 173
10000 55 71 89 100 123

 

[CJ]

It is just a huge toroid being wound, kinda lame.

But here are some A L vs Turns number for a 5 Band Pultec EQ:

A L Calculator

Formula: N = 1000* Sq Root (L/A L)
(L in milliHenries)

A L Turns 30 mH 50 mH 80 mH 100 mH 1 50 mH

100 = 548 707 894 1000 1225
200 = 387 500 632 707 866
300 = 316 408 516 577 707
400 = 274 354 447 500 672
500 = 245 316 400 447 548
600 = 224 289 365 408 500
700 = 207 267 338 378 463
800 = 194 250 316 354 433
875 = 185 239 302 338 414
1000 = 173 224 283 316 387

2000 = 123 158 200 224 274
3000 = 100 129 163 183 224
4000 = 87 112 141 158 194
5000 = 78 100 127 141 173
10000 = 55 71 89 100 123

 

[khstudio]

A L - Turns 30 mH 50 mH 80 mH 100 mH 1 50 mH

875 - 185 239 302 338 414

The 875 is the closest to Pat & My Pultec Clone inductor...
BUT - it doesn't look right :?

My first TAP @ 32mH was around 610 Turns.

 

[ioaudio]

if you get 32mH @ 610 turns, the Al value of the your core is 85,9

 

[khstudio]

[quote author="ioaudio"]if you get 32mH @ 610 turns, the Al value of the your core is 85,9[/quote]

OK... i think i get it... I'm using the "100 Turn" AL Value... like Pat did to get 876.
Yes it MAY not be perfect but it worked for both of us.

 

[CJ]

What AL value did your final turns come close to on the chart?
If you do not want to disclose that, no big.

Pat might mean 87.5 as his A L, not 875.

875 did seem more like a modern number, like a Fair Rite 78 series.

 

[ioaudio]

[quote author="CJ"]Pat might mean 87.5 as his A L, not 875.[/quote]

...which would calculate to 1309 turns for 150mh
edit:see below

 

[maxwall]

Assuming the truth has not been withheld with the original specs, I agree with Max (ioaudio), Kevin and Drpat.

my first tap at 27mH came in at nearly the same number around 610 - 640 turns. Which is not far off from 32mH.

I would not say it was so unless I wound a core first hand , which I did.

The math only gets you close , but not exact. exact turns will naturally slightly vary from core to core even if they are of the same composition. makes sense to me.

AL 875

Turns : 185 239 302 338 414

This is interesting data

Which number represents the first tap 27/32 mH ?

If I'm reading this correctly, I would guess 414 not 185 , but this must be for a core with a different composition than what I have.

there are three iron powder cores that I'm now aware of MPP, High Flux and Sendust, but endless recipes containing controlled percentages of Nickel, Iron, Silicon, etc etc each yielding different AL values and requiring different turns to get to desired taps using one Wire size like #34 as a starting point. When changing wire size then turns change and so does DCR as Max stated. this is assuming the physical dimensions of the core size remains the only constant.

 

[ioaudio]

lets do some math with pats data:

Wire Inductance DCR
Blue: 147mH 30.9
Green: 97.9mH 24
Yellow: 84.1mH 21.9
Orange: 50.9mH 16.7
Red: 32.2mH 13

wire used #34 (which has a DCR of 0,88368 ohm/meter)

next we calculate the average turn length of the core

core is about quarter sized, lets take this 24mm one
http://www.spulen.com/shop/product_info.php?products_id=332

external diameter = 23,6mm
inside diameter = 13,4mm
height=8,4mm

so one starting turn is 5,1 + 5,1 + 8,4 + 8,4 = 27mm
judging from the pics the core was half filled
ending turn at half: 8,45 + 8,45 + 8,4 + 8,4 = 33,7mm

so average turn length is 30,35mm

lets take the biggest tap which has 30.9ohm

30,9ohm resistance #34 wire = 30,9*0,88368 = 27,305712 meter

27,305712meter/0,03035meter = 899,7 turns for 147mH

this figure is likely too high - if the inductor was wound by hand, loose without consistent tension the avarage turn length was maybe larger -and the turns count smaller - but not a lot, maybe 50 turns less.

so the al value is
850 turns 147mH = 207.6

which is surprisingly low.


http://www.pacecomponents.co.uk/es/calculator.htm

 

[CJ]

ok i lifted this off the net



"AL Value (nH/N2)

The inductance rating of a core in nanohenries per turn squared based on a peak AC flux density of 10 gauss (1 millitesla) at a frequency of 10 KHz. Note: 35.0 nH/N2 =350µH for 100 turns =35.0 mH for 1000 turns."


Quarter size might mean the opening, not the OD.
Have to ask Dr Pat.


30 ohms is about right for 1300 turns of #36 on a 1 by 1.5 toroid core.

Here are some old numbers for a Fair Rite 77 material I wound:

27 mH - 92 turns
33 mH - 102t
47 mH - 122t
68 mH - 147t
82 mH - 161t
150 mH =218t

 

[khstudio]

[quote author="ioaudio"]
...which would calculate to 1309 turns for 150mh - if the core is sized like a quarter, its unlikely to achieve 30 ohm cause you would need a thinner wire.

edit:see below[/quote]

My 150mH Tap is just over 1300 turns & measures 31 ohm, using 34g wire.

... so, it's NOT unlikely.

#2 - I can't answer for Pat on the 876 AL. I also don't understand everything there is to know about this stuff & just followed his advise & instructions.

The cores I wound match the original Inductance Values & Resistance!

Wire Inductance DCR
Blue: 147mH 30.9
Green: 97.9mH 24
Yellow: 84.1mH 21.9
Orange: 50.9mH 16.7
Red: 32.2mH 13

Wound on a toroid core....

_________________
-Patrick

this is the info he sent me:

Quote:

As far as the Pultec inductor goes, here's the secret recipe:

The core that I used is black.
The dimensions are as follows:
Outside diameter: 23mm
Inside Diameter: 14mm
Height: 8mm

AL value of the core is 876µH. Test this by wrapping 100 turns around your core, then measure the inductance.

The wire gauge is #34AWG

Calculated turns are as follows:

32.2mH 606 turns
50.9mH add 156 extra turns
84.1mH add 218 extra turns
98mH add 78 extra turns
147.5mH add 240 extra turns

Turns formula:
N=100 x SQR(L/AL)

Where:
L= desired inductance in µH
AL= AL value in µH

So, for the 32.2mH tap...
N= 100 x SQR(32200/876)
N= 100 x SQR(36.75799)
N= 100 x 6.06284
N= 606.284 turns

 

[CJ]

ok excellent

he has his units in a different form, but its the same numbers.

it is indeed about the size of a stack of five quarters, wow, thats small.
the Lang Torwico is 25 mm by 10 with wire, so that adds up to the 5668 being probably of the same family/


More A L:

"Initial Permeability

Published values of initial permeability have been calculated from measured inductance values at 5 gauss on toroids (OD=.870, ID=.540, HT=.250) using Wayne-Kerr model 6425 or model 3245 LCR meters. Flux density and permeability were both calculated using effective core set parameters (Le, Ae and Ve) calculated per MMPA Toroid Standard No. FTC410. "

So they use a set magnetic path length, 5 gauss rather than 10 as a test level, so everybody has there standard.

Inductors are non linear anyway, so it is hard to make constants out of non constant variables.

But, if you are checking samples all day, it makes sense.
You use the number as a guide point for your QC samples.

 

[CJ]

OK this is easy,

A L = nH/N^2.

milli, micro, nano, 9 places...

A L = 1,000,000,000 (H/N^2) (converted my units to Henries)

Pats Inductor:

876 uH = .000876 Henries (converted Pat into Henry, sorry Pat)
and he used 100 turns to get that value,
we now have ditched the pesky units:

Al = 1,000,000,000 (.000876/100^2)

A L = 1,000,000,000 (.000876/10,000)

bring out the 10,000 so we can divide:

A L = 1,000,000,000/10,000 * (.000876)

A L = 100,000 * .000876 = 87.6 Bingo!

That is our A L number.

Now we can get turns:

30 mH = 585 turns
50 mH = 756
80 mH = 956
100 mH = 1069
150 mH = 1308

So thats the way it was back then, there best audio band powder core could do 87.6.

Now days that number is a joke.
But the 87.6 can also be used in a Ham Radio, it is so fast, those small, light di poles all bundled up in epoxy...

 

[khstudio]

Hey that's cool CJ... good work. :thumb:

Those #'s look good.

So the REAL AL # is 87.6. :!: :?:

This should help people (possibly) find cores that are close... having the correct AL. (1000 turns as opposed to 100)

On another note... There was NO way Pat could have got 1000 turns "Over" the original winds of his "Real" Pultec inductor... nor was he (or anyone in there right mind) going to ripe apart a "Classic".

........On the other hand, you might. :razz:

Thanks again to Pat for getting us this close :!: :thumb:
& CJ for digging deeper. :green: :guinness: :guinness: :guinness:

 

[gyraf]

Very good info indeed!

Mollypermalloy, AL81, 40mm dia: http://www.spulen.com/shop/product_info.php?products_id=963

Jakob E.

 

[khstudio]

[quote author="gyraf"]Very good info indeed!

Mollypermalloy, AL81, 40mm dia: http://www.spulen.com/shop/product_info.php?products_id=963

Jakob E.[/quote]

Now were getting some attention. :green:

I wish I could read German though.

OK... CJ, is it AL = 87.6 nH or uH ???
On the link Jacob posted it says "µ=60, AL-Wert 81nH".
... what the heck is a Wert?

Obviously I don't need any cores... just trying to help out.

 

[gyraf]

AL is most often measured in nH/(sqr-turns)

Also the above, if I'm not mistaken.

Jakob E.

 

[maxwall]

 

[CJ]

Hey Pat, there were some old Torwico cat pages floatin around here a while back I think.
Do you have any thing?

Thanks!

What is the flux?
We need the cross section, say it is 8mm square after the epoxy has been removed, thats 64 square mm, but the B Max formula needs square centimeters.
Convert square mm to sq cm with the internet converter, or rough it out, 1 mm = 0.1 cm, so 0.1 * 0.1 = 0.01

so Multiply by 0.01 to get sq cm:

64 sq mm * 0.01 = 0.64 sq cm.

Check by changing 8 mm to 0.8 cm >

0.8 * 0.8 = 0.64

OK, so B max = 10,000,000 V/4.44 f A(cm^2)N

So we need turns, a voltage level for a sine wave on the inductor, and a frequency at which the tap operates.


Lets pick the lowest frequency since this will put the most voltage on the coil. This will also be the tap with all the turns, which will make it more accurate due to the best linkage.

So choose 150 mH and 3 kHz, which is a typical Pultec value.
150 mH = 1308 turns.

Leave voltage out so we can build a B max chart for different levels:

B max = 10,000,000 V/(4.44* 3,000 Hertz * 1308 Turns * 0.64 sq cm )

B Max = 10,000,000 V/11,750,438 = 0.85 V

So if 1 volt of 3,000 hertz hit the core, it will see

0.85 * 1 Volt = 0.85 K Gauss.

At 10 volts, that would be 8.5 K Gauss, which is a lot for a powdered core.
Grain steel takes 16 to 18.
Permalloy 5 to 8.

Powdered cores would be even less.
That is because the domains are a lot smaller, they can not gang up like in a steel core.


So what is the flux allowed for that Torwico alloy?
Maybe it is in a spec sheet somewhere.