Problems with Coil Calculators

[nigelk]

I enjoyed, the book, - great work! There appear to be a couple of problems with the code and/or formula at the end of the book for the coil calculators. First off, the formula given for coil thickness is sqrt(N*w) whereas it appears it should be sqrt(N)*w ... this change makes the coil thickness output with the sample numbers on page 259 match the expected result of 3.76mm and the mean radius also matches. Note this error is in all three source formats. I still can't get the 2.741mH result however (I get 2.595mH) ... not a huge discrepancy but I haven't figured out where the problem lies. I've inserted all of the values from my (Excel) implementation below.

Inner Radius	216	mm
Wire Thickness	0.56	mm
Number of Turns	45
c	3.7565942
a	217.878297
x	7.4319E-05
S1	0.00027379
S2	2025
S3	0.55443273
S4	0.50000619
S5	5.03198936
S6	0.85001486
S7	4.68198068
Mean Radius	217.878297	mm
Coil Thickness	3.7565942	mm
Inductance	2.59584335	mH

[Qiaozhi]

Thanks for the feedback. This definitely looks like an error in Appendix A.

However, the Coil Calculator listed here on the forum ->
http://www.geotech1.com/forums/showt...oil-Calculator
does produce the correct result.

If you examine the JaveScript code for Version 2.0, it shows: c = Math.sqrt(numberOfTurns) * wireThickness;
which (as you pointed out) is the correct formula.

I fixed this problem in 2009, but unfortunately it looks like I may have published the code for Version 1.0 instead of 2.0 by mistake.
http://www.geotech1.com/forums/showt...5126#post95126

Note the problem with Log-base10 instead of natural logs. This could be why you're not getting the correct result.
If you spot any other problems, please let me know.

[nigelk]

Thanks for replying. I did some more digging based on the code you linked above and now understand at least where the differences arise from. First off, the code on Pg 262 also has two more problems: s7 = s4 + s5 - s6 should be s7 = s4 * s5 - s6. Secondly, the calculation of s6 on pages 261 and also on page 262 is 0.85 + 0.2 * x but as used in the code you just linked, the signs of the 0.85 and the 0.2 * x should be opposite ( i.e. to match your linked code, it should be s6 = 0.85 - 0.2 * x). Finally, as you mention above, there is confusion about Log10 vs. LN(). The code above uses LN() as do the VB and Tcl/TK code while the JS code goes out of its way to apply Log10. Can you confirm that LN() is correct? - otherwise the linked example above (and the expected results) aren't correct. Thanks again!

[Qiaozhi]

Thanks for replying. I did some more digging based on the code you linked above and now understand at least where the differences arise from. First off, the code on Pg 262 also has two more problems: s7 = s4 + s5 - s6 should be s7 = s4 * s5 - s6. Secondly, the calculation of s6 on pages 261 and also on page 262 is 0.85 + 0.2 * x but as used in the code you just linked, the signs of the 0.85 and the 0.2 * x should be opposite ( i.e. to match your linked code, it should be s6 = 0.85 - 0.2 * x). Finally, as you mention above, there is confusion about Log10 vs. LN(). The code above uses LN() as do the VB and Tcl/TK code while the JS code goes out of its way to apply Log10. Can you confirm that LN() is correct? - otherwise the linked example above (and the expected results) aren't correct. Thanks again!

The equation in Appendix A (Eq A-2) is the full [expanded] version of Brook's equation.

In the computer program: , so that the equation is reduced to:



Forget about the code listed on page 262 (and the VB and Tcl/TK versions), as these are Version 1.0 of the Coil Calculator that contains at least a couple of bugs. The calculator available for download at the top of the Coils Forum is version 2.0, which matches Eq A-2 in Appendix A.

Note that log() is the natural logarithm syntax for JavaScript, and that c=l and a=r in the code (when referring to the full equation).

[Qiaozhi]

Here are my hand calculations:

[nigelk]

Perfect, -thanks!