Calculating turns for a pickup

[meropx]

Hi, this is my first post, I'm learning about winding pickups and i have a few questions

The first and i think is the most important is:

If i want to play with the impedance of my homebrew pickups i need to know the turns it's need, thinking about it a few time i
have a idea to get it:

1 - Calculate the length per turn

-For Square end core:
Length_turn = 2*(width_coil + length_coil)

-For round end core:
Length_turn = Pi*width_coil + 2*length_coil

2 - Calculate the length for the wire
With the data of the wire, we could start to calculate the length of the wire we need to get the output impedance.

Total_length = "impedance wanted" / "Resistance in ohms per (km, foot, inch, mm, ...)"

If you use km the Total_length will be in km too

3 - Calculate the turns
Knowing the length of the wire and the length per turn it will be very easy.

Turns = Total_length / Length_turn

Use the same units for the both lengths

But this idea have a "little" problem, it will be the maximum turns, because it use the minimum length per turn, in a computer i could write a recursive function who calculate it with precision, but i want to lean to get the closer turns without using my computer

I'm thinking to get the wire's diameter(with the insulate) and with the coil's height i could know the max turns per layer, and get the number of layers it will take for the turns we need. Using this we could calculate the max core length, and using it to obtain the turns and calculate the average of both.

I'm on the right way? or it's the worst method to calculate the turns for the pickup?

Thanks for your time, and sorry for my bad English.

[Joe Gwinn]

There is a related procedure in http://home.comcast.net/~joegwinn/ that may be helpful.

[Sock Puppet]

There is also Adam Ziegler's software to try:

http://pickups.myonlinesite.com/programs.php

S.

[Steve Conner]

I've seen methods like this used in transformer design. They use the mean turn length, which I guess amounts to the same thing as Joe Gwinn's suggested algorithm.

You are correct in thinking that the calculation is recursive, because the value that Joe calls "Lmax" in his formula for mean length is a function of the coil dimensions, which you won't know until you've decided how many turns to put on, which you can't decide until you know Lmax, etc...

You may be able to use algebra to work it out, ending up with a formula that tells you the wire length as a function of the bobbin dimensions, wire gauge and number of turns. A mathematician would call this rearranging an implicit formula into an explicit one.

In practice though, I'd just guess at the height of the finished coil to produce an initial guess for Lmax. You can always compute Lmax again from the actual number of turns once you're done, and then redo the calculation once again with this value. If you're feeling really nerdy, you can repeat this until you "converge", and you'll have got the numerical solution, the same as your computer program would have done. (The algebra approach would have got you the analytic solution, but not all math problems have one, and sometimes you can't even tell if they have one or not.)

But the result can never be more accurate than your estimate of the fill factor, so it's hardly worth it IMO.

[Joe Gwinn]

I've seen methods like this used in transformer design. They use the mean turn length, which I guess amounts to the same thing as Joe Gwinn's suggested algorithm.

More or less.

You are correct in thinking that the calculation is recursive, because the value that Joe calls "Lmax" in his formula for mean length is a function of the coil dimensions, which you won't know until you've decided how many turns to put on, which you can't decide until you know Lmax, etc...

You may be able to use algebra to work it out, ending up with a formula that tells you the wire length as a function of the bobbin dimensions, wire gauge and number of turns. A mathematician would call this rearranging an implicit formula into an explicit one.

In practice though, I'd just guess at the height of the finished coil to produce an initial guess for Lmax. You can always compute Lmax again from the actual number of turns once you're done, and then redo the calculation once again with this value. If you're feeling really nerdy, you can repeat this until you "converge", and you'll have got the numerical solution, the same as your computer program would have done. (The algebra approach would have got you the analytic solution, but not all math problems have one, and sometimes you can't even tell if they have one or not.)

The fastest manual algorithm is Newton's Method (yes, invented by Mr. Gravity). What is simpler and also works very quickly is to make two trial solutions and interpolate (or extrapolate) to generate the next guess. This will converge within two or three cycles.

But the result can never be more accurate than your estimate of the fill factor, so it's hardly worth it IMO.

True enough.

[meropx]

There is also Adam Ziegler's software to try:

http://pickups.myonlinesite.com/programs.php

I will test it tomorrow in the laptop of my sister, when i use it with wine i need to kill it because it get frizzed (Wine, it's a software who allow to a Linux's user run some windows software).

There is a related procedure in http://home.comcast.net/~joegwinn/ that may be helpful.

Thanks Joe Gwinn, i have compared the both methods, and mine are a bit bigger (more or less a 7% more turns ) so i will use your method I will get again the Math's book of the university (omg. in 10 days i have the first exam )

But i will made a similar program like the "Pickup Calculator" but who works in Linux too.

Thanks for your replies, and see you the next day.

-------------------------------------------------------
Thanks from Spain.

[Spence]

Adam Ziegler's software

I've used Adam Ziegler's software a few times. It's pretty good. I find it most useful when trying to figure out an undetermined DCR for an unusual wrecked pickup that's in for repair.

[DrStrangelove]

Ziegler's Online pickup calculator

I will test it tomorrow in the laptop of my sister, when i use it with wine i need to kill it because it get frizzed (Wine, it's a software who allow to a Linux's user run some windows software).

Try the online version. You only need a web browser to run it.
http://pickups.myonlinesite.com/pickup.php

-drh

[David Schwab]

I can't get that online version to do much at all. It draws out the bobbin, and that's about it.

I'll try the Windows version in Virtual PC (I'm on Mac OS X).

[corduroyew]

when I try to run the windows version it says I haven't connected the requiered hardware.

[DrStrangelove]

Zeigler's pickup program

when I try to run the windows version it says I haven't connected the requiered hardware.

The Windoze application also counts turns if you have the necessary reed switch on the serial port. It is probably looking for that part. For more info, see
http://pickups.myonlinesite.com/serial.php

David:

The online calculator has several screens you must wade through.
It is a PHP script and should run on any browser written in the last 5 years.
Starting at http://pickups.myonlinesite.com/pickup.php --

First, you specify inches or metric, and click SUBMIT to move to the next screen.

Then you fill in the bobbin sizes. For a generic Strat bobbin, I use:

Core Length: 2.25
Core Width: .1875 (3/16" - the magnet rod diameter)
Core Height: .436
Bobbin Width: .625 (this is the width of the forbon flat)
Bobbin Thickness: .0625 (1/16" - the thickness of your average chunk of Forbon)
Round ends (default bobbin shape)

Click SUBMIT.
This gives you the empty bobbin picture.
For #42 gauge wire, enter

Wire Diam: .0025
Insulation Diam .0028 (assumes single build insulation)

Click SUBMIT.

The next screen lets you enter DC resistance or winds count
so you can make the final estimate.

Note that this calculator does not figure in a fill factor.
Bobbin is also misspelled as "Bobin".

-drh

[David Schwab]

David:

The online calculator has several screens you must wade through.

<snip>

The next screen lets you enter DC resistance or winds count
so you can make the final estimate.

Yep, I do al that, and the last part is as far as I get, and it does nothing when I click the button. I've tried wind count, DC resistance, etc.

It does draw the bobbin in, but no coil, and no results.

I tried it once last year and it worked.

It is a PHP script and should run on any browser written in the last 5 years.

Oh I know... I tried both Safari and Firefox and get the same results.

[Possum]

it works here...

I tried the instructions and it works, the big flaw in this program is it won't let you simply put in the wire O.D., how the hell are we supposed to know the bare wire O.D. of any wire we are using? I tried putting in a false number and the ohms results changes. Everybody's wire is going to be inconsistent size so you'd have to use some number thats not real off some wire gauge chart, so your results will be off. I think he should have just used wire O.D., I don't see the need for bare copper diameter, I don't follow that logic.....how the heck do you measure bare wire when stripping it will take some copper with it...

[David Schwab]

So it's not 42 AWG plus the insulation? That's what I was entering.

[meropx]

So it's not 42 AWG plus the insulation? That's what I was entering.

The "Insulation Diam:" is the diameter of the wire with the insulation, in the example http://pickups.myonlinesite.com/pick...=ohms&avo=6836 you could see it, Wire Diam: 0.0022 Insulation Diam: 0.0026 so the thick of the insulation is 0.0002