I have a few pieces of ETD 49 cores and I want to use them in a power supply. Input voltage will be 24V DC, output voltage should be 320 V DC, power as much as possible with this core. Switching frequency should be 100 KHz. Can anybody make the calculations for this transformer step-by-step so I can understand the details about designing such a power supply? Electronic part I already have - popular 3525 PWM chip and push-pull power stage with MOSFET transistors.
TNX
1/ Look at Siemens and Philips application notes. I remember them having some design examples, step by step. Try and search their webpages.
2/ There is are good books by Abraham Pressman and Keith Billings with a lot of practical examples.
If you really get stuck I can do the maths for you step by step when I have some spare time. At 100kHz there is a lot of things to consider, like layout, packing of the windings, stray inductance, type of core material etc.
It is difficult for a beginner to easily achieve a good working design even with all the data provided as a lot of the success depends on the construction of the transformer.
If you really get stuck I can do the maths for you step by step when I have some spare time. At 100kHz there is a lot of things to consider, like layout, packing of the windings, stray inductance, type of core material etc.
It is difficult for a beginner to easily achieve a good working design even with all the data provided as a lot of the success depends on the construction of the transformer.
I studiet a lot of literature and I made a few transformers, but I am not shure, that tht is it. I was not able to achieve the required power level or there was not enough place for wires or ... So, I am not shure if I did it in the best way. So, if you can do the math, it would be very helpful.
TNX!
I will see what I can do after the XMAS period when the wife stops giving me projects in and around the house!
I need to know a few things:
1/ What is the type of core material you have? N27, N30 etc. That will set the max operating freq and expected losses.
2/ Is the core un-gapped? Should be
3/ Do you do any current sensing in your design? I assume you are using a push-pull cct.
Maybe you can email me your cct to have a look at in the mean time?
1. Material: N67 Siferrit
2. Core ETD 49 with coilformer, ungapped
3. Yes, there is a resistor (one for both transistor assemblies) for current sensing. Configuration is push-pull.
4. The schematic is hand-made on a piece of paper and my scanner is dead so now I can not sent you the circuit but it is standard from application notes from ST except I made a stronger driver for MOSFET transistors.
About your wife, I have the same problem :wink: :wink: :wink:
On the same issue, I wish to ask the following question:
Has anybody tried salvaging used transformer cores for rewinding ???
( a good source would be old computer power supplies.)
I have tried using Terpentine with some luck, also I have tried heating
the core, also with some luck. I believe there must be a way to do it,
after that I think it is worth doing some experimentation.
I have seen many ferrit core calculations from TI and from RS (UK).
Like I have mentioned before, this will give you the correct values to use but if you don't take care in the construction of the actual choke or transformer you will have problems. It is like RF. You can't just connect everything together as per the diagram without proper placing and layout and expect it to work. With switchmodes you can have high peak currents and when you combine that with high leakage inductances in a sloppy construction, you can expect also very dangerous voltage spikes to be present on the switching devices. It takes practice to be able to construct a good transformer.
Inductor chokes should be constructed with a gapped-core or a powdered iron core to prevent saturation.
I have not checked the program too much for bugs, but if you find a problem PM me and I will fix it.
Later when I have more time I will include more topologies.
I`ve disassembled many ferrite transformers for repair in microwawe owen. Just put a shortcut to all pins ( to avoid HV induction ), put it into mw-owen and switch on. Open the door every 3-4sec and try pull out the ferrite with two pliers. Attention,very hot!