Although I've found many threads on OPTs and sound quality of tube amplifiers, no one seems to have asked this question. Power transformers (PTs) can have core different configurations (EI, C, torroid, etc), but core materials used in PTs appear less diverse than OPTs. However, which configuration do you think give lowest hum/noise? Obviously hum/noise affects sound quality, but do PTs affect sound quality in other ways? Am I correct to assume PTs do not affect sound as drastically as OPTs?
Any recommended manufacturers? I looked into Hashimoto who use EI cores, but they are quite expensive ($400+) for KT88/6550 amps.
Hashimoto Products
Technical Aspects
Plitron makes torroid PTs and was recommend by some.
Not sure if I should have posted this under Power Supply forum. If this is not the right place, I will post there.
Any recommended manufacturers? I looked into Hashimoto who use EI cores, but they are quite expensive ($400+) for KT88/6550 amps.
Hashimoto Products
Technical Aspects
Plitron makes torroid PTs and was recommend by some.
Not sure if I should have posted this under Power Supply forum. If this is not the right place, I will post there.
Last edited:
It depends on PSU design and build quality (e.g. grounding details). The better the design and build, the less will power transformer details affect the sound. Ideally, no effect at all. This is for hi-fi; guitar amps are different as there a 'weak' PSU may be part of the desired 'sound'.
Not as drastically, but the difference is pretty substantial. And it matters not whether it powers SS or tubes. The annoying thing is there doesn't seem to be a very obvious correlation to core materials or geometry; some just sound a lot better.
I have been asking this question for ages and although there is a lot of hand waving, nobody seems to really know. And unlike caps/resistors there is not a lot of empirical info on the sound of various brands. Some time ago i bought a few R-cores from VT4C which i really seem to like soundwise.
if you ask poorly formed questions like what does this circuit/topology/part "sound like" that is the case
the increasing resolution of ADC in electronic instrumentation across the boards has given rise to publication of lots of EMI design info for those pursuing how to build equipment that delivers their best resolution
if you have data on conducted and radiated EMI in your application environment, an experienced analog instrumentation design EE can design to a set of requirements for immunity/rejection, verify the results with domain appropriate measurement tools - it is just thorough application of well documented engineering principles
the increasing resolution of ADC in electronic instrumentation across the boards has given rise to publication of lots of EMI design info for those pursuing how to build equipment that delivers their best resolution
if you have data on conducted and radiated EMI in your application environment, an experienced analog instrumentation design EE can design to a set of requirements for immunity/rejection, verify the results with domain appropriate measurement tools - it is just thorough application of well documented engineering principles
All true, but I would go so far as to say that most audio kit is pretty horrible from an immunity and rejection perspective, and some is horrible from a radiated emissions perspective as well!
The audio world largely still thinks that it deals with DC to 20KHz, even when it is using parts with 100ps edge rates, in a room full of cell phones and WIFI, and still thinks that star grounding between the high speed and near DC areas is a good idea (Hint, a ground plane on a buried layer works better in almost all cases, the current will flow as as to minimise loop areas simply because of the physics).
As to power magnetics impacting the sound, only a issue if there is something else wrong somewhere in my experience, and the answer is usually to fix whatever else is wrong before worring about the core in low frequency power magnetics.
Regards, Dan.
The audio world largely still thinks that it deals with DC to 20KHz, even when it is using parts with 100ps edge rates, in a room full of cell phones and WIFI, and still thinks that star grounding between the high speed and near DC areas is a good idea (Hint, a ground plane on a buried layer works better in almost all cases, the current will flow as as to minimise loop areas simply because of the physics).
As to power magnetics impacting the sound, only a issue if there is something else wrong somewhere in my experience, and the answer is usually to fix whatever else is wrong before worring about the core in low frequency power magnetics.
Regards, Dan.
Just from personal experience (whatever that's worth), the best builds I've heard used C-core power transformers. No idea why, but toroid in the same spot didn't sound as good.
Good, an empirical observation, now measure both cases to find out what has changed then we will have some data to examine.
Yea, I know, but there is far too much voodoo in audio, numbers and scope pictures put the magic on a sound scientific basis (The music is still magical, but I really don't see that the gear needs to be).
Regards, Dan.
Yea, I know, but there is far too much voodoo in audio, numbers and scope pictures put the magic on a sound scientific basis (The music is still magical, but I really don't see that the gear needs to be).
Regards, Dan.
I'd be on it in a minute, if I had any of those C-Core any more. Alas, I do not. A worthwhile project tho, for sure. Maybe someone here does.
Would you like an answer, then?
Trouble is, the kind of people that believe in peripheral components impacting sound quality are the same type of people who are very good at mentally ignoring things, or confabulating them to suit their viewpoint.
Short answer: No. Not at all.
Qualified answer: as long as the power supply delivers the required voltage and current, with sufficient stability and low enough ripple to satisfy the amplifier's PSRR and voltage swing, then there will be no effect whatsoever.
These characteristics are most important in circuits with very little PSRR (SE and triode amps), and almost meaningless for those with large PSRR (you barely need any filtering at all on an SS amp, if you don't expect to use any voltage swing!).
Ripple and impedance becomes especially apparent on large signals, which bring the amp to clipping. Ripple causes clipping distortion and heavy IMD, while poor supply impedance causes weak transients and reduced power output. Outside of these conditions, conventional SS amps will see no difference.
Tim
how true, these same people ask questions just to reinforce their beliefs, and when they see answers contrary to their belief system, they run tantrums....😀
unsolicited advice, skip on the mundorf's or the v-caps and invest in really good power traffos like the Hashimotos......
conducted line immunity requires design, testing - certainly shows up when you start building instrumentation with 16 bit ADC and really look at the output
ground loop noise magnitude can depend on xfmr/winding/shield(if any) construction, termination
toroids early were associated with "high quality" in audio due to the expense, ability to make more compact designs - now with automated winding they are commodity parts competing for least material cost
and you can easily measure 5x or more inter-winding C with toroid vs split bobbin - so with attention to mag shielding or just more space in your amp you can get much better line isolation with EI split bobbin line xfmr
ground loop noise magnitude can depend on xfmr/winding/shield(if any) construction, termination
toroids early were associated with "high quality" in audio due to the expense, ability to make more compact designs - now with automated winding they are commodity parts competing for least material cost
and you can easily measure 5x or more inter-winding C with toroid vs split bobbin - so with attention to mag shielding or just more space in your amp you can get much better line isolation with EI split bobbin line xfmr
I think all C-cores would be air gapped by construction, and thus more 'resistant' to main DC 'noise'
and I expect most would also be more carefully built
some toroids looks like crap really
cheap and fast manufactured
not all ofcourse, but some I have seen looked that way
some years ago ordinary IE trafos were cheap, and toroids were expencive
but today its the opposite
funny thing is, our opinion or 'perception' also seem to have changed to the oppossite
we still want most what we cant have
I've seen R-Core in some recent Chinese gear. Better isolation?
R-core transformer-DIBAO Transformer
R-core transformer-DIBAO Transformer
"semi-toroidal" split bobbin is intermediate in pri-sec C, external field radiation - may be a good compormise in some cases
Do you mean nobody seems to know what you already prefer? 😀
When I choose transformer, I keep in mind maximal power that the amp draws. And calculate 10% drop and overheating for the max power. However, the majority of designers calculate for some imaginable "nominal power", but nobody knows what is the ratio between nominal and maximal powers, so I always assume it 1:1 -- it is one of secrets of high-end quality of P-P amps in class AB.
i DIY my own traffos so i am not limited by commercial offerings...i can make them as big as i want.......but how i wished that this was next door to me...Silicon Steel C-Cores / Goss C-Cores for 60 Hz and 50/60 Hz Frequencies
- Status
- Not open for further replies.