Toroid Output Transformer for Bass Amp

[CJ]

here is how you load the core,

 

[CJ]

you can use gorilla tape to hold the junction together,

we tried different methods to control the wire, foam iserts that5 fir into the rim to hold the wire, but they pop out from wire tension, so we cut up some nylon and made a small sqaure that snaps into the rim recess, that stays in place but slides all over the place, maybe some foam glued to the nylon insert would work, we might try some tweaks,

there is physics used on a modern coil winder, the wire spins like a lasso and centrifugal force keeps it where it should be, there needs tpo be a loop in the wire as the shuttle comes back from it's farthest point away from the core, there is nothing to pock up the slack so they throw the wire with speed, that is why a commercial machine has to have a controller, they need to star up fast,

good news is somebody saw this weird winder and offered up some tips on cheap machines in the area, 

 

[CJ]

we can still hand wind this thing as we only are doing one,

first we camfer the core to take of the razor sharp edges that might eventually nick thru the insulation and short the winding to the core, or short two or more turns so we have a shorted turn of low resistance wire which can be a bummer,

if you buy a small spool of wire you can trim the edges of the spool so you do not have to transfer wire to a stick, just pass the spool thru,

we have 63 feet of #16,

11 bucks free ship on eveilbay,

 

[CJ]

we used a layer of kraft paper under some glass cloth tape to protect the wire,

insl is 200 degrees, so you can not solder the leads, but this big wire is easy to scrape with a razor without damaging the wire,

we will not use the white pvc core caps as they move the wire too far away which means more leakage,

 

[CJ]

we did make use of the core holder table, which makes life easier,

high tension should be used to wind this big wire, you want it to stay close to the core for max linkage,

that area next to the core is like a  set of magnets when you try to hold them apart, the tinner the gap, the more force, same thing with transformers, as the wire gets close to the core, linkage goes up in a quadratic fashion,

in fact trip wound core makers will de-rate specs for varnish of epoxy coated cores as you can not get the wire close anymore,

 

[CJ]

turn length estimate turned out good, 100 turns wound, with enuff wire left for about 1.5 turns,

hands are stiff, can't play bass now, 

white tape will anchor a wad of turns so the leads don't work loose,

we do the secondary first so the big wire and heavy tension do not compromise the #29 that is about to go on,

and no crossovers like a commercial job,

 

[CJ]

100 turns is easy to work with as far as math for turns ratios and impedance ratios, and should provide plenty of Henries to drive the voice coils,

in fact, we can measure sec inductance before the primary goes on,

DCR came out to 0.2 ohms which is par for the course on most 8 ohm winds,

 

[CJ]

only one sig gen that we know of that will drive that speaker wind ,  HP 200CD, there might be a pair of 6L6's or similar in there doing the hard labor, even so, we only get 4 volts out of it at 5 Hz, voltage will go up as frequency goes up as XL is a finction of f,

this project tore the bench up, need clean up on aisle 5!

we measure the voltage across the coil and divide by the amps in the feed line to get ohms,

or reactance for the prop heads,  ;D

from that we divide by 2 pi and f to get L in Henries,

looks like a T1454 lurking left, R&D for Chameleon,

 

[CJ]

crankin pretty good, mucho Henries for an 8 ohm Sec,

if we wind 20:1,  100T x 20 = 2000 Turns,    turns 8 ohms onto  20^2 = 400 x 8 = 3200,

Primary Henries will be 1037 H at 20 Hz , similar to the Plitron featured up post, same ratio, same DCR,

never done toroid flux, guess you just multiply height by width to get cross section and apply Farady,

should get us close, 1" x 2.5" = 2.5  square inches,  16.129 Metric, which is what we need,

root 2.5 = 1.58, so sq stack of 150 EI would be the equivalent core,

100,000,000 E/ 4.44 x 20 Hz x 16 cm^2 x 2000T = 35 E,      18 K Gauss saturation/35 = 514 volts,

so the amp will start sounding good on about 7,  speakers get  514/20=25.7/8=3.2 amps,
3.2^2=10.33 x 8 = 82 watts,

need a pair of 6550 tubes and a EV SRO 15 maybe,

 

[gyraf]

thanks..

 

[CJ]

might be able to buy the ring shuttle from a winding machine vendor, which would have the little gadget that holds the wire and moves around the ring as wire is de-reeled,

meanwhile, we are going to tape up the secondary with kraft paper which will soak up varnish which will keep the winds from buzzing, do not want the wires to break,

 

[CJ]

add a poly layer and then wind, using a transformer bobbin for wire, put extra on,

 

[CJ]

split the primaries around the core as an experiment in voltage gradient distribution

lower the C,  extend freq range,

hook up leads,

20:1, 1000 turns each pri, measured series inductance, down from estimated 1000H to 425H why?  :-\

oh yeah, we run the pri at 500 volts, generator puts out 26 volts, so when we get the level up, inductance will come up as we climb the BH curve, there is a little loss due to the distance from wire to core since we wound the pri over the sec, so maybe we get 900H at 500 volts which is still plenty,

ratio check showed one winding high, unwrap turns till voltage splits equal, this will help keep the DC balanced in the core,

glass tape sticks better with heat, yellow tape can loosen,  dont want lead wires to pull out under stress,

 

[CJ]

whats for dinner?

howabout transformer tuna casserole and PBR?

might be a little tough, but will provide you with the recommended daily requirements  of iron and copper,

bake at 250 for 2 hours, smother with varnish sauce, re-bake and serve warm,  ;D

found the casserole dish in the neighbor's yard, toaster oven at Goodwill for $9.95,

 

[CJ]

here is a freak graph after varnish bake out,

not too bad, peaks out 3 times, 20KHz could be better but this is for a bass amp,

will hook this thing into a 6550 chassis and kick out the jams,

maybe do a side by side test with a 175 EI Peerless OPT,

 

[gyraf]

20KHz could be better

Whaddyamean?? That sweep is near-perfect! And it was probably even done without negative feedback?

Great story, this one...

Jakob E.

 

[Gus]

So when will the amp test happen?

Very Cool build

 

[CJ]

we finally bolted this transformer to the chassis and hooked it up today and it sounds awesome!

only problem is finding a speaker box that will handle the low end,

it has that thump action that makes the voice coils jump out of their sockets,  ;D

we are running a pair of 6550 tubes with cathode bias and about 460 on the plates,

there is a big drop in power when you go from fixed to cathode bias, but the sound is sweet, the transformer seems to make up for the lack of power as we never got past 5 on the dial as to save the speakers, lots of headroom and a smooth sub low response, can't wait to try it with an SVT cab or similar,
plenty of high end for snapin and popin, but that is for the pro bass geeks, we like the treble all the way down and the bass all the way up, Portaflex circuit preamp, octal preamp tubes, 6SN7 works well in the inverter spot, 6SL7 is stock but we have plenty of gain without it,

 

[CJ]

indoor/outdoor nonflamable

been thru a few incarnations, improves cooling,

tribal shirt cut up for front panel, Zulu print, ebay,

6550 outputs,  tried KT 120 , sounded nice til they blew up, Tungsol, wtf, over?