Hi tube experts.
I've picked up someone else's abandoned DIY project, to build a pair of mullard 5-20 monos. The circuit itself I'm fine with, I've built preamplifiers & phono stages before. But this is the first time I've had to deal with OPTs, so can anyone offer me some guidance?
What should I look out for when choosing one? Is "more-expensive-the-better" the rule here? Is size & weight the key, & can I get better results more cheaply just by oversizing? Can anyone recommend an OPT for this circuit?
Toroidals seem to have very high bandwidth but I'm keen to keep the project vintage-like unless there is a major advantage.
Also having looked at these things on the web for a day or two now I'm getting confused. Anode to anode load for t/x's supposedly designed for mullard 5-20 range from 3.5k to 6.6k. I would've expected it to be a constant. But I'm a newbie, so I know nothing
Thanks a lot
Ed
I've picked up someone else's abandoned DIY project, to build a pair of mullard 5-20 monos. The circuit itself I'm fine with, I've built preamplifiers & phono stages before. But this is the first time I've had to deal with OPTs, so can anyone offer me some guidance?
What should I look out for when choosing one? Is "more-expensive-the-better" the rule here? Is size & weight the key, & can I get better results more cheaply just by oversizing? Can anyone recommend an OPT for this circuit?
Toroidals seem to have very high bandwidth but I'm keen to keep the project vintage-like unless there is a major advantage.
Also having looked at these things on the web for a day or two now I'm getting confused. Anode to anode load for t/x's supposedly designed for mullard 5-20 range from 3.5k to 6.6k. I would've expected it to be a constant. But I'm a newbie, so I know nothing

Thanks a lot
Ed
Nominal 6.6ka-a is correct for an authentic Mullard 5-20.
Because toroids are virtually ungapped they only manage wide bandwidth if there is no out-of-balance DC current flowing in their core. Any out-of-balance current and you can kiss your LF performance goodbye.
In this instance, yes, bigger is generally better because it implies a lower core flux density, which lowers LF distortion. A good transformer will have many interleaved sections and will probably have four secondary sections that you connect in different ways to achieve the required matching. A secondary with taps is not likely to be as good.
You have to get the output transformer right - it's the heart of the amplifier.
Because toroids are virtually ungapped they only manage wide bandwidth if there is no out-of-balance DC current flowing in their core. Any out-of-balance current and you can kiss your LF performance goodbye.
In this instance, yes, bigger is generally better because it implies a lower core flux density, which lowers LF distortion. A good transformer will have many interleaved sections and will probably have four secondary sections that you connect in different ways to achieve the required matching. A secondary with taps is not likely to be as good.
You have to get the output transformer right - it's the heart of the amplifier.
- Status
- Not open for further replies.