ampage
Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

Re: Iteration is not required to find turns count


 :
6/3/2005 9:16 AM
anonymous Re: Iteration is not required to find turns count
Most of my tests on known pickups hit the reported windings at around 62% padding of the wire diameter. For the Barden Tele, I got exactly 2409 winds.  
 
"Houston, we have confirmation."  
 
For the Barden Tele, calculated Lmax is probably very close to the actual Lmax for a window .115x.230". With fabric tape, the half-coil thickness is .125", so your work is also a confirmation of the .010" tape dimension. Now we know something new.  
 
Dare I suggest that there is no tape on the blade core?  
 
What the iterative/quadratic method tells us is by how much these pickups diverge from maximal winds packing. It does not explicitly require the actual coil dimension, nor does uniquely it yield it -- only suggestions. A .115" thickness would imply 47 layers deep in a square lattice, but a 62% diameter padding puts it near 27 layers.  
 
"Reverse engineering" has become a nearly- derogatory term for a legitimate process. When a manufacturor is unable to support his product, this sort of specification reconstruction is easily justified.  
 
 
 
Congrats. We now have two more complementary tools.  
 
-drh  
--
 
Today
ampage
archive
And now, a word from our sponsors:


 
6/3/2005 11:25 AM
Huey Jackson

Have you considered adding mass as a factor to this equation? It seems that if you're trying to derive winding numbers from an unknown pickup, you might just be able to cut off the wire, weigh it and get a mass-per-turn measurement. NB: I don't understand about 90 % of what you guys are saying here.
 
6/3/2005 4:42 PM
Dr. Strangelove
Forensic pickupology: weigh the coil
Huey Jackson wrote:
quote:
"Have you considered adding mass as a factor to this equation?[/QUOTE]No, but it makes good sense.[QUOTE]It seems that if you're trying to derive winding numbers from an unknown pickup, you might just be able to cut off the wire, weigh it and get a mass-per-turn measurement."
Sounds like a fine idea for spec'ing a dead pup. Just for perspective, quick dirty calcs say that:  
 
Since MWS' wire tables say that #42 single builds are 50940 feet/lb.,  
-- the average 5" wind is just under 4 milligrams;  
-- a 7600 wind strat coil would weigh around 28 grams;  
-- a 4000 wind PAF coil would weigh around 15 grams.  
 
You'd need to weigh accurately in the 1 ounce region; a scale in the 50 to 200 gram range could do the job.  
 
I'd want a scale with 5 milligram resolution but 10mg (=.01g) resolution is cheaper, under $50 on eBay.  
quote:
"NB: I don't understand about 90 % of what you guys are saying here."
Not to worry.  
 
We don't understand all that much about actually building pickups. :)  
 
That's where you can help.  
 
If you've got a live pickup and you know the dimensions of its bobbin, you could give us some reality checks.  
 
-drh  
--
 
6/4/2005 8:22 AM
Joe Gwinn

On 6/3/2005 10:42 PM, Dr. Strangelove said:  
[QUOTE]Huey Jackson wrote: "Have you considered adding mass as a factor to this equation?" No, but it makes good sense. "It seems that if you're trying to derive winding numbers from an unknown pickup, you might just be able to cut off the wire, weigh it and get a mass-per-turn measurement."  
 
Sounds like a fine idea for spec'ing a dead pup.[/QUOTE]Agree. It's a good idea. This should work.  
 
It may be useful to remove the wax. This can be done by putting the rat's-nest of wire into a small pot of boiling water. The wire will sink to the bottom, and the wax will rise to the top. When the pot cools, the wax crust can be broken free and removed and the wire fished out, rinsed in denatured alcohol (to extract the water quickly; optional), allowed to dry on some paper towels, and weighed.  
 
[QUOTE]Just for perspective, quick dirty calcs say that:  
 
Since MWS' wire tables say that #42 single builds are 50940 feet/lb.,  
-- the average 5" wind [turn] is just under 4 milligrams;  
-- a 7600 wind strat coil would weigh around 28 grams;  
-- a 4000 wind PAF coil would weigh around 15 grams.  
 
You'd need to weigh accurately in the 1 ounce region; a scale in the 50 to 200 gram range could do the job.  
 
I'd want a scale with 5 milligram resolution but 10mg (=.01g) resolution is cheaper, under $50 on eBay.[/QUOTE]There is probably no point in accuracy and resolution much exceeding 1% of the actual measured value, so we can estimate the absolute resolution requirements of the scale. For the PAF example, 15 grams, 1% is 0.15 grams, so even 0.1 gram is fine, and 0.01 grams is wonderful.  
 
As a cross check, let's use the data from the Barden blade pickup used as the example for computation of turns count: The length of the winding was 1,040.6 feet of single-build AWG #43 wire, which is 66140 feet per pound, so the unwaxed winding weighs 1040.6/66140= 0.015733 pounds, or (0.015733)(453.6)= 7.1366 grams. One percent of this is (7.14)(0.01)= 0.071 grams, which is very close to 0.1 gram, so 0.1 gram resolution would work, and 0.01 gram resolution would be wonderful.  
 
I was hoping that a postal scale would work, but postal scales are good to no better than 0.5 to 1.0 gram, which is too coarse.  
 
Then there is the matter of accuracy. Many scales have high resolution (ability to discern small changes), but are not particularly accurate (ability to compare to a recognized standard). One red flag is if the scale specs talk only of resolution, and/or say "not legal for trade" on them.  
 
It would be useful to get a ~50-gram weight standard as well, with which to calibrate the scale. Having two calibration masses, say 25 and 50 grams, would allow use of an uncalibrated scale (so long as it's stable) to measure mass with nearly the same accuracy as the calibration weights. Caslibration weights are usually cheaper than good scales, and being solid lumps of brass, with no moving parts to break or lose, weights are going to be pretty safe to buy on eBay.
 
6/4/2005 12:45 PM
Dr. Strangelove

Joe Gwinn wrote:
quote:
"As a cross check, let's use the data from the Barden blade pickup used as the example for computation of turns count: The length of the winding was 1,040.6 feet of single-build AWG #43 wire, which is 66140 feet per pound, so the unwaxed winding weighs 1040.6/66140= 0.015733 pounds, or (0.015733)(453.6)= 7.1366 grams. One percent of this is (7.14)(0.01)= 0.071 grams, which is very close to 0.1 gram, so 0.1 gram resolution would work, and 0.01 gram resolution would be wonderful."
A 5" turn of #43 weighs 2.8mg.  
±71mg corresponds to a ±25 wind uncertainty;  
±100mg means ±35 winds.  
 
I would prefer at least a ±.05 gram accuracy here.  
 
The measurement depends on good technique and clean wire.  
 
My lab experience suggests boiling in slightly soapy water, rinse with distilled water and scrupulous drying before weighing four times (one to discard as an outlier).  
 
-drh  
--
 
6/4/2005 2:54 PM
Joe Gwinn

On 6/4/2005 6:45 PM, Dr. Strangelove said:  
[QUOTE]Joe Gwinn wrote: "As a cross check, let's use the data from the Barden blade pickup used as the example for computation of turns count: The length of the winding was 1,040.6 feet of single-build AWG #43 wire, which is 66140 feet per pound, so the unwaxed winding weighs 1040.6/66140= 0.015733 pounds, or (0.015733)(453.6)= 7.1366 grams. One percent of this is (7.14)(0.01)= 0.071 grams, which is very close to 0.1 gram, so 0.1 gram resolution would work, and 0.01 gram resolution would be wonderful."  
 
A 5" turn of #43 weighs 2.8mg.  
 
±71mg corresponds to a ±25 wind uncertainty;  
 
±100mg means ±35 winds.  
 
I would prefer at least a ±0.05 gram accuracy here.  
 
The measurement depends on good technique and clean wire.[/QUOTE]The reason I suggest 1% as the goal is that no matter how careful we are, or how accurate our scales, the achievable accuracy is limited by manufacturing variation in magnet wire and wire coating thicknesses, as well as the amount of stretch during winding. So, we are not really leaving any accuracy on the table by going for a scale resolution between 0.1 gram and 0.01 gram.  
 
But there is no harm in greater resolution, and a scale can be used for many things. The real tradeoff is accuracy, precision, and price; pick any two. Accuracy is compliance to a recognized standard. Precision is range divided by resolution, and thus is the number of different weight values the scale can report.  

 
quote:
"My lab experience suggests boiling in slightly soapy water, rinse with distilled water and scrupulous drying before weighing four times (one to discard as an outlier)."
I haven't tried soapy water, but it sounds like a good idea. I would use a detergent intended for hand dishwashing.  
 
As for drying, I would suggest rinsing with denatured alcohol after the distilled-water rinse, as the alcohol will remove the water quite efficiently, without allowing the minerals in the water to end up on all surfaces, and itself evaporates quite rapidly.

 
6/4/2005 3:27 PM
Dr. Strangelove

Joe Gwinn wrote:
quote:
"As for drying, I would suggest rinsing with denatured alcohol after the distilled-water rinse."
It's not glassware.  
 
Unless you absolutely know you won't encounter decaying formvar coatings or plain enamel, avoid organic solvents. Old pickups use these insulations.  
 
A magwire insulation that merely crazes on solvent exposure has partly dissolved.  
 
The part about scrupulous drying needs elaboration.  
 
My inclination is 1hr @ 100F since some of the coatings are hygroscopic enough that the manufacturors recommend some kind of drying before potting/varnishing a wound product.  
 
-drh  
--
 

<<First Page<PrevPage 2 of 5 Next> Last Page>>