Two 4Ω secondaries.

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

TerryG

Active member
Joined
Sep 18, 2021
Messages
27
Location
California USA
I think I know the answer to this but just in case, and because I am a noob I'll ask anyway.

What is the best way to wind two 4Ω secondaries? I was thinking if I had two reels of wire I could just wind them at the same time, or wind one on top of the other making sure to keep track of their orientation. Also, would this give me an 8Ω if I externally wired them in series, properly oriented of course.

Also if I have two 4Ω windings, then I am imaging like a power transformers primaries using either 120vac or 240ac would be similar. I would have the option of secondary impedance of 2Ω, 4Ω, 8Ω of speaker impedances depending on how I wound them.

That brings up another question. For some reason, I start off by using the color of tape the same as the color of wire I am going to solder. But the end of the winding I start with the first section of 1/4 of the primary winding I mark with red tape, and the start of the secondary winding which I start with white, as on an amplifier schematic they are on the same side of the represented core. I suppose I should be checking with my dual trace oscilloscope if my windings are in phase or out of phase by 180 degrees. The only reason this is meaningful is for the binding post for two channels to be in phase with each other. Is this how you guys do this?

Thanks in advance!
 
Also, would this give me an 8Ω if I externally wired them in series, properly oriented of course.
No, two 4Ω windings in series will give a 16Ω winding. Just calculate the secondary numbers for a given primary winding number and you will see it.
Also switching them in parallel won't give you a 2Ω winding. It is effectively still the same number of turns, just two in parallel so half the resistance, but nominal impedance stays the same.

If what you want is 2Ω, 4Ω, 8Ω outputs, calculate the correct numbers and make it one winding with taps for 2Ω and 4Ω, no switching necessary.
 
Last edited:
That brings up another question. For some reason, I start off by using the color of tape the same as the color of wire I am going to solder. But the end of the winding I start with the first section of 1/4 of the primary winding I mark with red tape, and the start of the secondary winding which I start with white, as on an amplifier schematic they are on the same side of the represented core. I suppose I should be checking with my dual trace oscilloscope if my windings are in phase or out of phase by 180 degrees. The only reason this is meaningful is for the binding post for two channels to be in phase with each other. Is this how you guys do this?
I must admit I don't really understand what you mean, but are you sandwiching sections? If not, I would suggest "two wires in hand" (otherwise known as bifilar), then you use an ohmmeter to identify which is which.
 
Quick question along this line. So if two 4Ω windings in series will give you 16Ω, what will two parallel windings do for you? Say two 8Ω windings, how does this reflect upon the primary winding?
 
I must admit I don't really understand what you mean, but are you sandwiching sections? If not, I would suggest "two wires in hand" (otherwise known as bifilar), then you use an ohmmeter to identify which is which.
What I am asking is how do you guys keep track of your windings in a vacuum tube output transformers. When I start I start with the primary winding, so I mark the magnet wire with RED tape because I am going to solder the begining of the primary with a red wire. The end of the primary I will solder a blue wire on that goes to the tubes plate. On the secondary I use a white wire soldered at its magnet wire, when I am done with the secondary I will solder a yellow wire so I mark that piece of magnet wire tail with yellow tape. This is how I keep track of which magnet wire gets soldered to which color of wire. I basically use the common standard I have found for all the output transformers for tubes amplifiers I have ever seen. I supposed I could write a note on a piece of masking tape to mark my wires as to which gets soldered to which color.

I suppose I could use all the same color wire, and use different colors of shrink wrap conforming to the standard to tell me which wire is which.

Some of the things I do I feel are common sense, but I don't know how much of a standard there is with some things. For example I am putting kapton tape between windings to reduce any electrical arching interaction between windings. I also paint on some poly
 
Volker already answered that. Read again.
Ok so the nominal impedance stays the same, but will lessen the resistance.

I noticed on a transformer I took apart to use the core and bobbin its secondary has a bifilar wound two secondaries. The nominal impedance was 8Ω, but it was a quandry to me why they would do that. Was it simply to possibly get the current rating they desired but not have the gauge wire that the math told them they needed? I have learned a lot by taking professionally made transformers apart to scavenge the parts. Until I get a stable supplier for cores and bobbins this will be my life. I am getting quite the barrel of scrap copper though.
 
I noticed on a transformer I took apart to use the core and bobbin its secondary has a bifilar wound two secondaries. The nominal impedance was 8Ω, but it was a quandry to me why they would do that.
Were the two sections hard-wired in parallels?

Was it simply to possibly get the current rating they desired but not have the gauge wire that the math told them they needed?
It's one of the possible reasons. Another would be that it allows a sligtly more taut winding.
 
Very often two strands of smaller gauge wire are easier to handle (less stiff) than an equivalent resistance single strand - and the bifilar will often pack the available bobbin space better (less wasted "air" space).
 
you can wind the two 4 ohm sections side by side, voltage will never be greater than about 50 volts, wire is good for 5 KV each,

this might make a more stable transformer (no oscillations).
 
Very often two strands of smaller gauge wire are easier to handle (less stiff) than an equivalent resistance single strand - and the bifilar will often pack the available bobbin space better (less wasted "air" space).

Has anyone ever tried to use square wire because square wire would pack better? And not just adjacent loops but different levels would sit flatter.

Also it might exhibit lower impedance for a given wire gauge since it would have a greater cross-section circumference to area ratio and the surface of wire is lower impedance compared to the core.

Meaning run it through a square(ish) die just before it reaches the bobbin?
 
Last edited:
Do you have a supplier for square wire? I have thought about this, but wouldn't it increase capacitance in an OPT?
AFAIK, in audio, only speaker manufacturers use square magnet wire. I visited the JBL factory eons ago, they had a machine that took round wire and made is square(ish), by rolling round wire through ball bearings. This contraption was designed and built by James B. himself.
Square wire is also used in high power xfmrs and motors, where stray capacitance is not a concern.
Square magnet wire seems to be commonly available, but only in large gauges.
https://mwswire.com/magnet-wire/square-magnet-wire/
 
rectangular wire is very common in hi-pwr transformers,

when you have a short on a power line, and if it happens at the peak of the sine wave, then you are going to have a lot of torque on the coils. so you need to have those coils stitched down pretty tight on the core legs, to do this, they use a core bracket system whereby you can tighten some nuts down in order to squeeze the coils down hard on the vertical plane. this is going to put pressure on the windings. if you use round wire, then the turns can overlap each other from the pressure. with rectangular wire, you have a flat face butting up against each other which tends to keep the coil solid while undergoing the torque down step during assembly.

there is another advantage to really hi current situations which use big rectangular wire, maybe 4 inches wide by 3/8" thick, you can drill it out for buss bar attachment, with big round wire, you would need to either use a silver solder connection or some type of clamp which can cause problems if the clamp gets loose with age.

we use to have to rub this powder onto the buss bars with a damp cloth, it was called cool -amp. it put a silver plating on the copper. and it also gave you carpal tunnel after doing a big three phase unit.
 
Twisting the two wires together with 1.5 turns per inch will give you a thicker wire that packs really well. Just think LITZ... IT's also a good way to prevent layers from cutting into earlier layers and creating capacitance issues, though maybe not for that low of an impedance.. Many old transformers that were dual, trick and quad filler were wound with ribbon wire, which explained the reason they were smooth sounding. When the supplier of that type of wire went away, transformers changed tone (late 70s) so I started winding them with the twist and it was the same effect.
 
Western Electric used about every trick in the book, including the ribbon wire when the turns count was low,

Pulse Transformers are very interesting, huge C cores with low turns and hi voltage. This one Nickel C Core which was about a foot high and 4 inches wide went for 2000 bucks a piece from National Arnold. They used to get pissed when the engineer sent it back because the damped sinusoidal was not quite right. The faces of the C core looked like a mirror from all the polishing. Crepe wrap around the wire and even the breakout leads, then submerged in Silica Oil which is super expensive compared to the usual mineral oil. Used a lot in the medical field.

I want the schematic for an MRI machine and maybe a CT Scanner, anybody?
 

Latest posts

Back
Top