If I wanted to use mosfets, I wouldn’t bother with the tubes in the first place 🙂. I have a nice power amp that has a bunch of mosfets already.
Nice! You are serious, you know too to use capacitors as a precaution to block the 100+V DC on the various parts of the amplifier. Also be careful with the ground, as long as you ground on the ground you should be alright, don't create loops inside the USB , it will blow up the computer supply and the dacs
I know this, but It’s always a good reminder 🙂 easy to get forgetful when your other gear can take the high voltage.
The computer I mostly use won’t talk to the USB oscilloscope for some reason, so that ends up running off of an older one that only gets used for the oscilloscope. Mostly I use the tek scope because it’s faster, and easier. But the usb job can do bode plots and such too, with the built in software, and its own built in generator. If I get bored with other projects, I’ll get Pete millet’s isolator/Adapter PCB for more safely connecting to PC soundcard.
These look interesting, but they don't have many options. You'd think with computer winding techniques, it wouldn't be hard to have a large variety made to order in relatively quick time frame. Surface shipping is less painful than I would have imagined for two units, and the price for the smaller ones isn't outrageous, but sadly they are only 6 ohm out.
http://softone.a.la9.jp/english/
http://softone.a.la9.jp/english/
In my readings, it seems like a lot of smaller operations cling to EI manufacturing not because it is actually superior, but because it costs too much to buy the machines needed to make toroids. Especially for the winding of toroidal transformers. I just saw a short video clip of how R-core transformers are wound, and it's absolutely brilliant. Split plastic bobbin that is placed onto a straight section of the core and driven with a rubber idler wheel. That would give many of the advantages of EI construction over toroids, with fewer of the downsides. I imagine a DIY winder could make a machine to wind these type of cores, bobbins could be 3D printed, or manufactured any number of relatively simple ways. Even making such a core wouldn't be an insurmountable task, since it's relatively easy to wind thin section material like that. The hardest part would be the shaping of the core to make it round in cross section.
Most toroid winder machines do random winds, which causes high distributed capacitance using many back and forth sweeps. More expensive toroid winders can do progressive winds, where the full winding is done in a chevron fashion, in a single pass around the core, these are needed for quality OTs.
The original Antek OTs had rather poor freq.response. The improved ones, with in between specs, may be using a bunch of sectioned random winds (limited angle section around core for each one), don't know, just guessing.
There are E-I techniques that can do similar high quality OT's, but you don't see them commercially. Commercial OTs invariably use "scrapless" laminations with short E's. One approach is using "long E" lams where the three E section parts are extra long. Leakage L goes down significantly when the air leakage path becomes long and equal length with the lamination path. The DIY winder can use "constant V transformer" laminations (resonant topology like SOLA) without the shunt pieces. These laminations can be found that will allow two standard bobbins side by side on the E. Very useful for split bobbin type P-P OT's where each P-P side can have symmetrical windup and balanced characteristics.
Then there are E -only- laminations (one Co. in NJ had a patent on these at one time). The back side of the E is twice as wide as normal, so no I lamination is needed. They would push the long E's all the way together to get the overlap with the next flipped lamination back side, so there were no "butt" gaps.
For SE OT'S, one can just use two sets of E facing each other across a spacer gap. Bobbin on each side.
Those Softone prices look like they are almost twice the Edcor prices. ??? The big advantage of toroid OTs is the continuous winding around the core minimizes leakage L. R cores and cut C cores with bobbins on each side loose that advantage, short leakage paths around the coils. One could make a long version of such cores for long bobbins, but I have never seen any.
Another thing that seems to get mentioned often is that a larger than needed OT would have high magnetic losses. This is not necessarily so. Magnetic losses vary as the square of excitation, so using a big transformer below it's usual magnetic excitation can have equal losses. Just expensive and heavy.
The original Antek OTs had rather poor freq.response. The improved ones, with in between specs, may be using a bunch of sectioned random winds (limited angle section around core for each one), don't know, just guessing.
There are E-I techniques that can do similar high quality OT's, but you don't see them commercially. Commercial OTs invariably use "scrapless" laminations with short E's. One approach is using "long E" lams where the three E section parts are extra long. Leakage L goes down significantly when the air leakage path becomes long and equal length with the lamination path. The DIY winder can use "constant V transformer" laminations (resonant topology like SOLA) without the shunt pieces. These laminations can be found that will allow two standard bobbins side by side on the E. Very useful for split bobbin type P-P OT's where each P-P side can have symmetrical windup and balanced characteristics.
Then there are E -only- laminations (one Co. in NJ had a patent on these at one time). The back side of the E is twice as wide as normal, so no I lamination is needed. They would push the long E's all the way together to get the overlap with the next flipped lamination back side, so there were no "butt" gaps.
For SE OT'S, one can just use two sets of E facing each other across a spacer gap. Bobbin on each side.
Those Softone prices look like they are almost twice the Edcor prices. ??? The big advantage of toroid OTs is the continuous winding around the core minimizes leakage L. R cores and cut C cores with bobbins on each side loose that advantage, short leakage paths around the coils. One could make a long version of such cores for long bobbins, but I have never seen any.
Another thing that seems to get mentioned often is that a larger than needed OT would have high magnetic losses. This is not necessarily so. Magnetic losses vary as the square of excitation, so using a big transformer below it's usual magnetic excitation can have equal losses. Just expensive and heavy.
Last edited:
I would definitely propose this guy! https://www.ebay.com/itm/403210099697?hash=item5de131dbf1:g:kswAAOSwXFlgaY5Z
very good quality transformers from a very knowledgeable guy! Highly recommended from me as I have used a lot of his opts and chokes. Also he can listen to your very specific needs and design customizations and last but not least all done with excellent craftsmanship and very low prices. His personal e-mail is thermioniclab@gmail.com if contact him outside eBay mention my name and you are going to save even more on the prices.
very good quality transformers from a very knowledgeable guy! Highly recommended from me as I have used a lot of his opts and chokes. Also he can listen to your very specific needs and design customizations and last but not least all done with excellent craftsmanship and very low prices. His personal e-mail is thermioniclab@gmail.com if contact him outside eBay mention my name and you are going to save even more on the prices.
They are good, they are made with symmetrical windings. They make a huge difference in sound. If you know how to make the best use of them go for it. You need a stable circuit.
Unfortunately stereo, and in particular tube amplifiers, has become a niche market with many of the younger generation opting to have smart[hones and earbuds.
A number of well known manufacturers have called it quits and with deminished supplies prices are rising. Finale Audio has stated that they will only target the high end market. The prices for AirTight have also gone up considerably over the years (amongst other niche vendors).
The pandemic has had its own effect on pricing - we need only to have a look worldwide at what house prices have done. Chip shortages is doing its own thing to the car prices and computing gear so no surprise on parts prices. (unless you want to use Chinese stuff with its own risk of not knowning its performance).
A number of well known manufacturers have called it quits and with deminished supplies prices are rising. Finale Audio has stated that they will only target the high end market. The prices for AirTight have also gone up considerably over the years (amongst other niche vendors).
The pandemic has had its own effect on pricing - we need only to have a look worldwide at what house prices have done. Chip shortages is doing its own thing to the car prices and computing gear so no surprise on parts prices. (unless you want to use Chinese stuff with its own risk of not knowning its performance).
For headphones or low power, I just noticed these Edcor 5 Watt SE OTs, CXSE for $30, with an extended freq. response ( 40 to 20K ). 5K and 7K primary versions. Cheaper than the GXSE ones. A 300B only makes 7 Watts.
Last edited:
6 ohm is perfect, usually people would connect 4 ohm outputs to 8 ohm speakers to get better damping, harder on the tube, 6 ohm is mostly what speakers are in modern world, and is a perfect balance between 4 and 8 ohms. There are other factors such as feedbacks which greatly change the ratio to the loudspeakers
Attachments
Not testet. They make trannies to spec., and prices seem to be on the right side.
http://www.ogonowski.eu/transformatory/transformatory-glosnikowe-se/
http://www.ogonowski.eu/transformatory/transformatory-glosnikowe-se/
There are cheap options, and a lot of members in this forum contributed into finding alternatives. Some have even used power toroidal transformers as output transformers. Ogonowski is a cheap source, Toroidy also is.
It's all relative.
What's the cost of filling up your car and the cost of a service?
I remember the prices of a tube and an output transformer in relation to someone's weekly salary long time ago. They've not been any cheaper than just before the pandemic started.
I've talked to that guy (Matt). Good guy, but very busy. He can wind custom taps if you give him time (lots of it). He's up front that most of his OT's are for guitar use. Link below of 40W thats more $$ then your link above lists response of 50 -15K... still pretty good for the $$, IMO
Jim
https://www.ebay.com/itm/154808554845?hash=item240b4f155d:g:mJkAAOSwEahhnYRw
have you looked at Cinemag.biz?
I bought some big output trafos from them for a prosound dac modification.
I don't know what specs you need for your project but take a look they have many options.
Also, I'm not sure if they are "affordable."
I bought some big output trafos from them for a prosound dac modification.
I don't know what specs you need for your project but take a look they have many options.
Also, I'm not sure if they are "affordable."
sadly those do not have UL taps on the primary. I'm going to start a new thread soon with info on what I'm building, once I've settled on the transformers.