Triad AO-3039T Output Transformer

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I was never able to wind a transformer of a card bobbin without wings, in the middle of the winding the wire starts to get loose on the side and everything crumbles together.

I'm looking forwards to see your technics to wind this type of bobbins

Thanks CJ
 
DC flux estimate, do not know the exact plate current for the output tube, #40 wire is rated at 10 ma, engineers know they can safely double that, hoping they did not but used about 16 ma for which we get about 4.3 Kilo Gauss, if they ran 20ma, that would explain the heat soaked look of the wire,

dropping the relative permeability on this chart down to 800 which is normal for a gapped core, ungapped perm would be about 5000 for non grain steel like we have here. So we have about 4 KG AC flux and 4 KG DC flux = 8 K Gauss total , well below the saturation point of non grain steel which is around 18 to 20 K Gauss.
dc flux.png
 
DC flux estimate, do not know the exact plate current for the output tube, #40 wire is rated at 10 ma, engineers know they can safely double that, hoping they did not but used about 16 ma for which we get about 4.3 Kilo Gauss, if they ran 20ma, that would explain the heat soaked look of the wire,

dropping the relative permeability on this chart down to 800 which is normal for a gapped core, ungapped perm would be about 5000 for non grain steel like we have here. So we have about 4 KG AC flux and 4 KG DC flux = 8 K Gauss total , well below the saturation point of non grain steel which is around 18 to 20 K Gauss.
View attachment 88425

On the current handling for AWG, I couldn't find a chart for anything over 40awg so here's a chart I made up bye eye-balling with my organic number cruncher filling in the gaps. I use this chart for my winds (but only for power amp OT and power transformers tho).
I will get around to winding an input transformer with some old '50s lams with the E's stored upwards so the mojo doesn't fall out for that DI box i need.

Doug
 

Attachments

  • AWG current handling.pdf
    200.5 KB · Views: 34
The Best of Rod Stewart, 7th pressing, 5 gram non-virgin vinyl album cover is about to take a hit.

i was in the record store yesterday and it was packed, elbow to elbow, like a super spreader party or something, and i am thinking the vinyl revolution had begun, which is a good thing. people have finally had enough of digitized music, console makers take heed,
sell your sony pro-tools stock now.

rs.JPG
 
Last edited:
why not make a 3 section coil into a 6 section coil to improve hi-end?
yes, we use a chambered bobbin,

capacitance decreases as a square of the number of chambers, so if we double the amount of chambers, we have 2^2 = 4 so we have 1/4th the capacitance.

a simple 7 step process will save you a Cosmo Corp 2000 dollar setup fee to have them do it, everything was done strictly eyeball>

chambered bobbin.jpg
 
first secondary done, using beefy #32, a worthy selection for a 600 out,

random wound? no layer insulation, wtf?

that's right.

what kind of a trashed out sound are you gonna get out of that?

i don't know, why don't you ask Rupert Neve, or Marinair, or St Ives, or Carnhill, or API, or Quad 8, or Telefunken or...s1.JPG
 
you can tape over a piece of Nomex to increase your insl level to 3000 volts on that primary wire about to get taped to the bobbin, Nomex 410 is good for 1000 volts per mil, so 30 Nomex good for 30 KV etc.

we are done with the 2900 turn #36 AWG secondary, looks like everything is gonna fit,

we are going do double up on the Nomex pri-sec insl to disrupt the symmetry of the coil. what this does is lower resonant peaks by de-tuning the coil, just like a tank circuit in a radio, this might be one of the reasons random wound coils sound so good, you breakup a little of that perfectness of having precise layers wrapped by the same insulation. also, capacitance is a function of the plate orientation, the most capacitance comes from parallel plates, anything else uses cosine to offset the calculation, so in a random wound coil, the wire crosses itself frequently, causing a slight angle between each turn which can add up in a 2900 turn coil like this.


pri.jpg
 
we now have 579 ohms primary dcr and 28 ohms secondary,

so we trimmed i do not know how many ohms off the primary because i could not measure it because that wind was open, if 850 is close then we shaved 271 ohms there and 34-28 = 6 ohms off the secondary.

Fill Factor came out alright, we might have gone 1 gauge bigger on the secondary wire next time, but no big deal, fill factor is the percentage of copper in the window. 90 percent is around normal.

put the leads on, slam the core in, start testing inductance and freq response, ratio test, no need for hi-pot with nylon bobbin and Nomex, dry out in toaster oven, dip in varnish, bake overnight, put on end bells, retest, ship out,

done winding.JPG
 
Ohh Damm,
I thought you were going to do it like the original and wind it in a cardboard bobbin with no wings!!!!!!!!!

o_Oo_O☹️☹️

I'm disappointed CJ, thats too easy

😁😆
 
i know, i almost feel guilty. i do the nice layers for the hi powered stuff because you have to.
the poly wire is so much better these days than the old enamel that you can get away with random winding with no problems.

hookup, do not have red, so orange for the B+ lead, Andy will have to deal with that.
yellow for the hot, purple for ground ,did not have blk and grn, Andy will have to deal with that.

wrap the un-tinned magnet wire around the lead wires first, then solder. the solder will melt through the insulation if you did not use 205 C wire. if you tin the magnet wire first, you will stress it out and it might fracture somewhere down the line.

i hand wound the secondary winds and used the stepper/traverse for the primary to keep the turns per layer the same.

i use a quick layer of yellow mylar to hold the magnet wire in place while i put on the main tape layers. that way, if you slip while tensioning the outer tape, it will not grab the magnet wire and rip it out of the coil which we have done on one occasion.

hookup.jpg
 
over to the lamination table. 87 EI. using 0.014" grain orient which has a slightly higher perm than the non grain 0.023" original iron, even with a gapped core the perm will come up a bit, thus boosting the bass.

warming up the old HP because it takes awhile to kick in on the 20 Hz range,

the antique generator is what gives the xfmr the Funk, understand?

lams.JPG
 
@CJ is there a difference in performance in a transformer between different types of magnet wire insulation? Does polysol act differently from plain enamel, from Formvar, etc?

could/do the different coatings impact the capacitance between layers of wire, due to film thickness, from different dielectric properties, through some other mechanism?

Or do you not notice a difference in practice?
 
pickup winders seem to be into experimenting with different wire,

i try to use older wire like Essex mainly because it seems to be of higher quality than the new stuff, which hopefully means a more dependable coil,

as far as new wire there does not seem to be a lot of choices out there, Remington, Elektrisola and that is about it unless you go for some off shore stuff on evilbay,

this transformer is doing about 110 Henries at 20 Hz and 20 V-rms which is decent,
freq response is flat out to about 50 K, there is a resonant peak at 240 KHz, which is pretty far out so we are happy. this would make a good transformer for the Bass DI that is floating around here.

calc: 20V/1.6 ma = 12,500 ohms Inductive Reactance (Xl)

use formula Xl=2 pi f H to get Henries, (f = 20 Hz )

to get inductance, we set the generator at 20 Hz, nobody cares about inductance at 20 K Hz because there isn't any,
ok so apply 20 volts and have an ammeter in series, in this case we get about 1.6ma of primary AC, this is known as the excitation voltage, kind of an idling current needed to jump start the xfmr into operation. assgasgrass nobody rides for free, same thing with transformers,

if you do not have any keeper lams like me,you can simply flip a pair of lams to hold the I bars, .

stk.JPG
 
Back
Top