Flux and field strength are not the same thing, offcoure H field goes up when you load the secundary wheras fluxdensity B (depending on the resistance of the windings) sligthly decreases.
In a OPT with M6 iron at 0.8T givs the highest inductance. I strive for at most 0.8T at 20Hz (guarantees 0.4T above 40Hz).Below 0.4T is something worth to strive fore. Why? Simply because distortion really starts to increase above levels of 0.4T wheras at levels below 0.4 T is does not decrease by much. For the same reason it is important to me to have the best possible balanced tubes in a PP amp, I would not be very happy to have things f...d up by a unnecessary addional Bdc. Keep Btot below 0.4T above 40hz (where most of the action is) and you are a step closer to have something worth listen too. This way, according to a guy that knew a thing or two bout transformers, additional transformer distortion will stay below 0.04%
Last edited:
I know at least Beyer used some true zero-airgap cores, does Lundahl too?might have been Lundahl
burnt traffo smells......when end-bells are removed, you can see the charred paper layers, even the coppers turned black....
I didnt know that dc from DMM could damage transformers, although i don't think that mine Are a zero air gap core...
I wonder, with a continuity test, testing Primary leads together should be open right? I ask that because i have same old trafo and doing continuity test some are buzzing and other don't..so i'm afraid to connect to 230v things shorted!
there's some minimun ohmic value i should read both on primary leads and secondary leads according to different output values? Ie 230v to 12-18-24v
Thanks for your patience, i'm new on the matter! I know, These are maybe stupid questions for you guys!
I wonder, with a continuity test, testing Primary leads together should be open right? I ask that because i have same old trafo and doing continuity test some are buzzing and other don't..so i'm afraid to connect to 230v things shorted!
there's some minimun ohmic value i should read both on primary leads and secondary leads according to different output values? Ie 230v to 12-18-24v
Thanks for your patience, i'm new on the matter! I know, These are maybe stupid questions for you guys!
You do not need to know the thickness of the lamination if you have a coil on the iron. If there is no coil you will have to wind one and then find the volts per turn that gives the highest no load impedance. The highest inductance is with M6 at around 0.8T. In a OPT use 0.8T as maximum at your lowest desired frequency. In a M6 mains transformer 0.8T gives best iron loss efficiency and low magneto striction (mechanical noise). It also limits the B to maximal 1.6T during switch on. It comes with a cost, relative copper losses go up, over all efficiency will go down. If the measured highest impedance is at at a higher calculated B than 0.8T then your iron is something better than M6, if lower than it is a lower quality iron. If you want the highest power for a given iron you can increase your V/turn up to a max of 2 times of the value where you measured the highest no load impedance. 2 times is the about the max, so with allowance for 10% over voltage this would be 1.46T at the nominal input voltage. With M6 iron 1.4-1.5T is what most professional transformer manufacturers would use in order to be able to compete with other products that are maxed out for power and overall efficiency. The above is for EI, UI and M iron in the sizes we use in audio. With a toroid or C core somewhat higher induction can be used.
Last edited:
Also, there is no such thing as "correct number of turns". There are only designer choices that have to be made. A transformer designer/manufacturer may choose the highest B he can get away with because he wants the highest power with as little costs as possible. On the other hand, Tony for instance, uses a comparable low B to make sure his transformers perform well in his audio projects.
I remember a case when a 10kVA transformer where "upgraded" to 16kVA just by stamping 3x400V (instead of the 3x380V he used earlier) and temperature class "F". Same iron, same coils, all that changed was a few pieces of higher temp class isolation sleeves and a new "licence plate" and price tag. The customer had to pay twice for this new invention, first the higher purchase and over the years the additional energy cost due to higher losses and he got also rewarded with lousier regulation and shorter life expectancy.
There is no correct number of turns, power rating a.s.o, only different choices that can be made
I remember a case when a 10kVA transformer where "upgraded" to 16kVA just by stamping 3x400V (instead of the 3x380V he used earlier) and temperature class "F". Same iron, same coils, all that changed was a few pieces of higher temp class isolation sleeves and a new "licence plate" and price tag. The customer had to pay twice for this new invention, first the higher purchase and over the years the additional energy cost due to higher losses and he got also rewarded with lousier regulation and shorter life expectancy.
There is no correct number of turns, power rating a.s.o, only different choices that can be made
Last edited:
i am glad that you made this comment, it is obvious that those who do not make transformers are unaware of this, i have been giving this hint for a long time now.........
... so say you want to wind an audio transformer and work out your number of turns initially and then decide afterwards you also want to provide some negative feedback say 10%.. so you would add in an extra 10% winding in the middle which will not affect the prim and sec voltages as originally calculated .. and will reduce the frequency response per the equations etc ?
the 10% extra winding is added to the total primary winding, assuming cathode feedback, frequency response is hardly affected if at all.....
The secondary impedance is given by the load, not the transformer.
The transformer only reflects this load into the primary. For example a ideal stepdown transformer with a winding ratio of 10:1 and terminated into a 8 ohm secundary load would transform this into 800ohm on the primary.
Adding 10% cathode fb-winding would change the winding ratio into 11:1.
With the same 8 ohm load the reflected primary impedanc would now become 11x11x8=968 ohm. The secondary voltage will drop. By how much depends not only on the winding ratio but also on the source impedance. Triodes will work happily into a 21% higher load. Pentodes can be a bit more grumpy. Make sure the additional winding is spread out over the full windingwidth.
Last edited:
tony,
I know you are not a world renowned expert on this stuff but you are one of the few [only ?] individuals here that actually winds things and provides photos so any suggestions are helpful .. so another few quick queries on our favourite subject.
I have most of the parts now to get a winder built hopefully next week but I am still trying to finish my first complete draft of this program and I remain confused about cathode feedback.
Say I was not using CF I would design as normal .. each trafo end terminal to plate and centre tap to B+.
So I have arrived at a design with 20 primary layers, a transformer designed for particular valves with a particular primary plate to plate impedance and therefore specific secondary impedances. Pat Turner shows this info using small tables on his site....which someday I hope to replicate!.
OK I change my mind and I decide to use CF instead, say 10% CF.
Now the big question is should I ADD two more windings to the design - and that would then affect my secondary impedances .. so they would need recalculation ?
OR would it be better to designate two of the existing windings to CF ?
If I designated two of these my secondary impedance table would then not need to be recalculated, is that right ?
I am also wondering if it matters where I would place the extra winding in the transformer OR if I used two of the existing layers, which two of the 20 I should use.
The reason I ask this is because I want the program to spit out the winding diagrams and I need to resolve these things in my own head before I program it, the visual bits of this [such as they are] will be time consuming for me to do as a non-programmer!!
Pat Turner on his output-trans-PP-calc-3 pages - Fig 23 - shows a plan with four secondaries and a CF winding, but if I had a trafo with an ODD number of secondaries would it matter exactly where I place the CF winding .
In other words should I use the two windings on either side of the central secondary or even a bit further out or does it even really matter that much exactly where I place the feedback windings ?
Again I know you are not a professional builder but if you have used CF in your designs you may be able to shed some light. I will probably only find this out properly when I wind one - and attempt to measure values.
I have a shed load more queries on practical things but this post would be way too long, I will leave them until I actually build the winder.
I will send you a copy of the program when I finish it if you want to give it a spin.
It is probably about twenty pages long so far and counting ...about 60% done.
I know you are not a world renowned expert on this stuff but you are one of the few [only ?] individuals here that actually winds things and provides photos so any suggestions are helpful .. so another few quick queries on our favourite subject.
I have most of the parts now to get a winder built hopefully next week but I am still trying to finish my first complete draft of this program and I remain confused about cathode feedback.
Say I was not using CF I would design as normal .. each trafo end terminal to plate and centre tap to B+.
So I have arrived at a design with 20 primary layers, a transformer designed for particular valves with a particular primary plate to plate impedance and therefore specific secondary impedances. Pat Turner shows this info using small tables on his site....which someday I hope to replicate!.
OK I change my mind and I decide to use CF instead, say 10% CF.
Now the big question is should I ADD two more windings to the design - and that would then affect my secondary impedances .. so they would need recalculation ?
OR would it be better to designate two of the existing windings to CF ?
If I designated two of these my secondary impedance table would then not need to be recalculated, is that right ?
I am also wondering if it matters where I would place the extra winding in the transformer OR if I used two of the existing layers, which two of the 20 I should use.
The reason I ask this is because I want the program to spit out the winding diagrams and I need to resolve these things in my own head before I program it, the visual bits of this [such as they are] will be time consuming for me to do as a non-programmer!!
Pat Turner on his output-trans-PP-calc-3 pages - Fig 23 - shows a plan with four secondaries and a CF winding, but if I had a trafo with an ODD number of secondaries would it matter exactly where I place the CF winding .
In other words should I use the two windings on either side of the central secondary or even a bit further out or does it even really matter that much exactly where I place the feedback windings ?
Again I know you are not a professional builder but if you have used CF in your designs you may be able to shed some light. I will probably only find this out properly when I wind one - and attempt to measure values.
I have a shed load more queries on practical things but this post would be way too long, I will leave them until I actually build the winder.
I will send you a copy of the program when I finish it if you want to give it a spin.
It is probably about twenty pages long so far and counting ...about 60% done.
way i understood it, any winding that does not connect to the speakers are primary windings it does not matter if it is connected to plate or to cathode.....
therefore, the turns ratio is for total primary windings to all secondary windings....
i think Pat Turner gave out all the information in his website, just use them....
i use the traditional pencil and paper technic in making my traffos....
there is a book by Wolpert available for download at YvesM's website...that book will give you an easier to follow design examples...http://www.diyaudio.com/forums/tubes-valves/193087-how-make-audio-output-transformer.html
post#6 even gives you a software to calculate/design an output transformer...
therefore, the turns ratio is for total primary windings to all secondary windings....
i think Pat Turner gave out all the information in his website, just use them....
i use the traditional pencil and paper technic in making my traffos....
there is a book by Wolpert available for download at YvesM's website...that book will give you an easier to follow design examples...http://www.diyaudio.com/forums/tubes-valves/193087-how-make-audio-output-transformer.html
post#6 even gives you a software to calculate/design an output transformer...
I have installed Yves and it is good but what I am aiming for is something that will do everything - apart from actually wind it.
I want to show the winding arrangements on screen including insulation, primaries and secondaries, also how the secondaries are divided up and organised...in series or parallel.
Type in valve parameters, voltages, number of pairs of valves and RUN, then wind the transformer basically by looking at the screen .. if you will.
Basically you could leave your pencil and paper in the drawer and never have to use them again.
Partly this is also a programming exercise for me - I am also intent on doing a program for the McLyman book *Transformer and Inductor Design* for switchmode stuff as well if I can get around to it.
I want to show the winding arrangements on screen including insulation, primaries and secondaries, also how the secondaries are divided up and organised...in series or parallel.
Type in valve parameters, voltages, number of pairs of valves and RUN, then wind the transformer basically by looking at the screen .. if you will.
Basically you could leave your pencil and paper in the drawer and never have to use them again.
Partly this is also a programming exercise for me - I am also intent on doing a program for the McLyman book *Transformer and Inductor Design* for switchmode stuff as well if I can get around to it.
no more than a few minutes, the actual hard work is in the actual building, planning, gathering the parts, etc......i am a practical guy, i know what i should be concerned with and what matters the most, if you think of everything you may never get started at all....
oh. another thing going for me, i know at least three old-timers in the business of rewinding tube OPT's, they give me lot's of tips, plus i get to see how they actually do their thing.... and then i get to dismantle opts from discarded tv sets, amps etc....