Audio Asylum Thread Printer
Get a view of an entire thread on one page
|For Sale Ads|
In Reply to: RE: " Maybe I should be looking for a toroidal transformer" posted by 1973shovel on June 10, 2017 at 11:33:15
I will only agree with what John Curl says if it is amended along these lines:
curl: What we forgot was that the windings of a toroid transformer are wound on top of each other, rather than in separate bobbins.
to make it factual: What we forgot was that the windings of a toroid transformer are typically wound on top of each other, rather than in separate bobbins.
I guess my point is that the problem curl mentions is not a function of the toroid but a function of how nearly all of them are wound. There is absolutely no reason why someone cannot wind a toroid with the primary as a semicircle on one half of the core and the secondary on the other half effectively giving the results of a dual bobbin device and I really have a problem with people lumping design choices choices and factual data into the same pile. It is not the fault of the toroid that it is wide bandwidth, it is the fault of the designer of the transformer. Next thing you know we will all be accepting the universal fact that tubes have a mushy vintage sound.
While I am at it, there is another gross overgeneralization used in magnetics wrt C-cores vs. EI cores and that is the following often repeated fact about the superiority of C-cores
The grain direction stays parallel to the winding length for the entire magnetic circuit for C-cores and stamped laminations always have some portion of the magnetic path that runs perpendicular to the grain.
The problem here is what happens when you are using a non-oriented material? As many of you know, I love nickel as a core material and it just so happens that the materials I use are all non-oriented and lets just say that being lectured by customers on the "universal superiority" of c-cores gets tiresome. To close things out I will give the one universal benefit of the tape wound cores is the ability to use much thinner lamination thickness than with stamped lams and the one benefit of stamped lams over C-cores is the possibility of a much smaller airgap.
" There is absolutely no reason why someone cannot wind a toroid with the primary as a semicircle on one half of the core and the secondary on the other half "
Dave, have you ever wound a transformer like that? My experiments with this technique at RF frequencies indicate that coupling degrades and core losses increase significantly. Perhaps things are different at the mains frequency, or maybe I did something wrong. I tried it only on one occasion using various cores and several hundred watts of RF. I quickly abandoned the idea in favor of isolating the primary and secondary using an insulating layer of fiberglass tape between them.
Buy Chinese. Bury freedom.
yes I have. wound in that manner and as you say the leakage inductance goes up which means regulation goes down. I was actually rather surprised on how much of a difference the pri-sec proximity effected both the output voltage and regulation at line voltages and I was using *gasp* Stamped M6 Laminations so I did have the cross grain issue which may have made things a bit worse. Of course this is a function of all transformers so anything you do to reduce winding to winding capacitance increases leakage so you just have to pick your poison so to speak.
Along the line of the toroids... Antek has a 50VA unit that will fit your needs for $28 and the single secondary winding will be fine for your bridge. I dropped a note to John @ antek about doing a 50VA version with dual secondaries for FWCT use to match his 100VA and 200VA units which are just too big for linestage / phono use.
"Antek has a 50VA unit that will fit your needs for $28"
Antek has several products that would work for this, including the 160V AS-05T160. I just wish these weren't made in China. Even Hammond is doing that now for some of their line. I'm having to source more and more NOS to be sure of getting Western-made components.
Buy Chinese. Bury freedom.
I have a few of those Hammonds (single primary version) and the secondary has a higher resistance than the primary, about 10%. I would guess that they're designed with about 10% step-up (about 132V unloaded) to give 120V output at the rated 0.25A load. In reverse, the voltage would be 10% low unloaded, dropping another 10% at full load. These could be used in the usual direction with a voltage doubler - that's how I expect to use them.
I used a 1:1 toroid to power a small amp (w/ voltage doubler), and wound an additional 6V secondary on it for heaters.
Buy Chinese. Bury freedom.
If you reverse connect a transformer you are running it outside of its design parameters. Give it a try with cheap transformers but bear in mind that winding current and voltage ratings may be compromised as may B-H characteristics.
"If you reverse connect a transformer you are running it outside of its design parameters."
Hmm... I'm not sure why you say that as a blanket statement. Of course with any transformer, whether reverse connected or connected in the standard way, it is always possible to run it outside of its design parameters. But if the transformer is reverse connected and the normal secondary (now being run as the primary) is not asked to pass any more current than it would when running as the secondary in the usual configuration, is there any reason why it would be "outside the design parameters"? (And assuming the voltages on the windings are essentially within the design voltages.)
Would the transformer even "know" that it was running in reverse-connected mode? All it knows is that a primary current within its design parameters is flowing, and a secondary current within its design parameters is flowing. Why would the transformer care?
Again, I am assuming both primary and secondary currents and voltages are kept adequately within design parameters. I think this was the case in TK's original example. (120V on a nominal 115V winding is within acceptable tolerances, I think.)
For power transformers I agree since we are just talking voltage and current however if we expand this to audio transformers then we also have to take into account bandwidth and frequency response doesn't always work as expected when you change connections.
Post a Message!
This post is made possible by the generous support of people like you and our sponsors: