I have a 50VA 2*12V 2.05A toroid mains transformer. I load the secondary with 50W power using load resistor.
The toroid starts to heat. After half an hour it cannot be touched by hand. I measured the temperature and the surface temperature seems to be 116 C 😱 The core inside the plastic tape must be much more hot.
Is this level of heat generation normal for a 50VA toroid with 50W load ?
The toroid starts to heat. After half an hour it cannot be touched by hand. I measured the temperature and the surface temperature seems to be 116 C 😱 The core inside the plastic tape must be much more hot.
Is this level of heat generation normal for a 50VA toroid with 50W load ?
An externally hosted image should be here but it was not working when we last tested it.
I think usual transformers are designed to run up to 100C. But you do use a heavy load, usually we don't go that far. Maybe up to 40-45W would be more reasonable.
By the way, how did you get that image? I need that method! 😛
By the way, how did you get that image? I need that method! 😛
Modern transformer materials can cope with heat. If it came from a reputable manufacturer then it should be OK up to its rated load, provided the ambient temperature is not too high.
Oooh that is a nice picture. With I could afford anything made by Flir.
Did you factor in derating for the type of rectification used? Double check my figures but I seem to remember its x1.7 for full wave+cap and .8 for half wave+cap.
Did you factor in derating for the type of rectification used? Double check my figures but I seem to remember its x1.7 for full wave+cap and .8 for half wave+cap.
That's not good at all. With a power factor of 1 as with resistor load you shouldn't be getting anywhere near that amount of heating with rated load. Just for reference, what is the power into the primary. In other words, how much is the transformer burning with 50W out? (Efficiency.) What is the power consumption with no load?
Any heat would concern me. Does it get hot without a load? If it does, shorted turns. If not poorly designed transformer.
70 degrees C is the absolute limit, I would say.
70 degrees C is the absolute limit, I would say.
Hi,
For typical music amplifiers its a fairly useless test.
Music amplifiers are not laboratory amplifiers producing continuous
RMS sine waves at full power. The optimum design of a transformer
for a good music amplifier is very different to a laboratory amplifier.
Sadly most this is obscured rather than expounded by manafactures.
rgds, sreten.
For typical music amplifiers its a fairly useless test.
Music amplifiers are not laboratory amplifiers producing continuous
RMS sine waves at full power. The optimum design of a transformer
for a good music amplifier is very different to a laboratory amplifier.
Sadly most this is obscured rather than expounded by manafactures.
rgds, sreten.
Hmm! A transformer will produce what the ratings state. End of story. They are designed and built for any requirement. If they were for a temporary load, say MIG welding, then yes, they will overheat but a transformer will deliver continuously what it is rated to do and in my experience does not get hot.
According to this web page about transformers:
Practical considerations : Transformers
transformer power ratings are broken out into "Classes". A summary of these is given in the following table taken from the web page:
You show 116C winding temperature. If the transformer is rated class F or H, then this would fall within spec. I am not familiar with these specs, I just thought I would bring them up for discussion. Not sure what type you are using, or into what class a typical audio power transformer would fall...
Also, it's possible that the transformer is rated for a particular duty cycle less then 100%...
I think it depends on what lifetime you want from this transformer under this load, and how often the PS will be required to deliver a constant 50W average power. This might depend on much compression has been used in the music program that the amp is reproducing (assuming that is the use for the transformer).
-Charlie
Practical considerations : Transformers
transformer power ratings are broken out into "Classes". A summary of these is given in the following table taken from the web page:
- Class A: No more than 55deg Celsius winding temperature rise, at 40 Celsius (maximum) ambient air temperature.
- Class B: No more than 80deg Celsius winding temperature rise, at 40 Celsius (maximum)ambient air temperature.
- Class F: No more than 115deg Celsius winding temperature rise, at 40 Celsius (maximum)ambient air temperature.
- Class H: No more than 150deg Celsius winding temperature rise, at 40 Celsius (maximum)ambient air temperature.
You show 116C winding temperature. If the transformer is rated class F or H, then this would fall within spec. I am not familiar with these specs, I just thought I would bring them up for discussion. Not sure what type you are using, or into what class a typical audio power transformer would fall...
Also, it's possible that the transformer is rated for a particular duty cycle less then 100%...
I think it depends on what lifetime you want from this transformer under this load, and how often the PS will be required to deliver a constant 50W average power. This might depend on much compression has been used in the music program that the amp is reproducing (assuming that is the use for the transformer).
-Charlie
In my opinion, anything that clears the boiling point of water on camera in a DIY audio application had better be a vacuum tube. Especially if you recently paid for it.
By the way, if there is a shorted turn it would be very easy to see with no load and such IR camera equipment.
By the way, if there is a shorted turn it would be very easy to see with no load and such IR camera equipment.
Last edited:
100 degrees C may well be within the design spec of the transformer but it certainly wont last very long.
I certainly would not be comfortable with a transformer running that hot in anything I designed. I tend to derate to no more than 70% of the rating and in higher current applications with toroids where noise is a concern I tend to derate to 50% - this keeps heat and noise down and efficiency up.
Are both secondaries wired in parallel? Using something like a 3 ohm load resistor?
I would be leery of even using it, having stressed it so.
For my gear (transformer) rise above ambient is generally <40C or better.
Are both secondaries wired in parallel? Using something like a 3 ohm load resistor?
I would be leery of even using it, having stressed it so.
For my gear (transformer) rise above ambient is generally <40C or better.
I agree. 50 degrees C is TOO hot for me. The transformer should not be too hot to touch - and ideally should be comfortably warm.
Last edited:
116°C at the surface is little indication of the worst case internal temperature.
Could it be 130, or 140, or 150°C, or a lot more?
Could it be 130, or 140, or 150°C, or a lot more?
A shorted turn might show as a concentration of heat. The pic shows fairly uniform heating.
Could this be a 60Hz transformer running on 50Hz? 220V on 230V? Or is it just Chinese?
Could this be a 60Hz transformer running on 50Hz? 220V on 230V? Or is it just Chinese?
The transformer is INDEL TST 50/004. Made in Poland (apparently). Nothing wrong according to the specifications on the label.
An externally hosted image should be here but it was not working when we last tested it.
This heat problem bothers me, so I made some additional temperature measurements (IR camera) with different load dissipations.
Here the transformer surface temperature [C] at different RMS load power dissipations:
It looks like, for example, if we want to keep the surface temperature below 50 C the load power cannot exceed 25 W. This would be only half of the rated power of 50 VA.
The test is done by connecting both secondaries in series, then to rectifier full bridge and a capacitor bank of 30000 uF.
The load resistor is selected to correspond the indicated RMS power. I measure the load current (true RMS), and the power is then P = R*Irms^2.
The level of voltages and currents are fine, so the transformer is functioning as it should. Only the heat . . .
I would feel stupid to put a fan for a transformer 😛
Here the transformer surface temperature [C] at different RMS load power dissipations:
It looks like, for example, if we want to keep the surface temperature below 50 C the load power cannot exceed 25 W. This would be only half of the rated power of 50 VA.
An externally hosted image should be here but it was not working when we last tested it.
The test is done by connecting both secondaries in series, then to rectifier full bridge and a capacitor bank of 30000 uF.
The load resistor is selected to correspond the indicated RMS power. I measure the load current (true RMS), and the power is then P = R*Irms^2.
The level of voltages and currents are fine, so the transformer is functioning as it should. Only the heat . . .
I would feel stupid to put a fan for a transformer 😛
- Status
- Not open for further replies.