PA transformer

Just one small point Arvido,

To "guestimate" this transformer and duplicate it, you need would really be well equipped if you have the impedances of the output from the driver stage and the input to PA stage.
If you do not know the impedances then you can look at the power level and the RMS voltages at the input and output of the transformer.

From the burned coil you should be able to have a rather good guess.

If it were mine then I would guess at something like this (assuming a 15:1 impedance ratio for a driver stage):

• 0.15mm Dia. 8+8 turns for the primary
  
• 0.40mm Dia. 2+2 turns for the secondary
  

Notes

• The 8+8 is two bits of wire twisted together and 8 turns wound on the former
  
• The 2+2 turns is wound the same way on top of the 8+8 input winding
  
• In this context, 1-turn is wire pushed through one hole and returned through the other hole
  

This should give you a practical starting point. This is how I would go about winding the coils.

Test it at 1.8MHz and 30MHz. If it gives a low output on 1.8MHz then you need to scale the number of turns up a bit.
The radio also has a form of AGC/ALC power control, so that if the coil is not exactly the same as the original, then the radio will make adjustments accordingly in operation.

• When you fire-up the radio, be sure to begin with the power level really low, just enough to test it
  
• Run the radio for several minutes at a low power level and check that it does not get hot or draw too much current
  

Well, that is my 2-penneth. I hope that you have luck with the transformer.

BR Harry - SM0VPO

Yes, but be careful not to burn the transistors out.

VBR from Ivan

John_1981 wrote:Is it possible to remove the burnt out windings and measure the length of the wire used? may not be as good as counting the turns but could give you a reasonable estimate. 

My guess is that the core is ferrite as its for broadband service. The parallel secondary may be to improve balance on the push pull output where a bifiliar winding is not possible.

Could the rewind be done by trial and error? start off with a high primary: secondary ratio then gradually reduce the ratio for maximum RF volts across R26//R27, or the expected power output at the aerial socket. 

If there are too few primary turns it wont work at the LF end and the ferrite will get hot. If it doesn't work well at the HF end there are too many primary turns. 

Is there anything about what voltages to expect at the test points? if so this will give a clue as to the impedances you need to work with.

Fine,"trial and error" could be a good solution.

Is it possible to remove the burnt out windings and measure the length of the wire used? may not be as good as counting the turns but could give you a reasonable estimate. 

My guess is that the core is ferrite as its for broadband service. The parallel secondary may be to improve balance on the push pull output where a bifiliar winding is not possible.

Could the rewind be done by trial and error? start off with a high primary: secondary ratio then gradually reduce the ratio for maximum RF volts across R26//R27, or the expected power output at the aerial socket. 

If there are too few primary turns it wont work at the LF end and the ferrite will get hot. If it doesn't work well at the HF end there are too many primary turns. 

Is there anything about what voltages to expect at the test points? if so this will give a clue as to the impedances you need to work with.

If the core is made of iron powder (carbonyl), it should keep its magnetic properties after cooling down. If it is of ferrite, it must be replaced.

I am confused by those two secondary windings in parralel. It probably has to do with the bandwidth, but I have no clue how to calculate the number of turns. Maybe someone else knows?

VBR from Ivan OK1SIP

Ivan wrote:Hi,
is it possible to dismantle the burned one and rewind it using new wires? Is the core O.K. or did it lose its magnetism due to overheating?

BR from Ivan

I have dismantled already,but it was so badly burned that no chance to count turns.Core is no broblem new one can be bought.

Hi,
is it possible to dismantle the burned one and rewind it using new wires? Is the core O.K. or did it lose its magnetism due to overheating?

BR from Ivan

I have a badly burned transformer for RACAL 2000 transceiver, need somebody to design this,not available as spare part.I add a schematic .This should work 2-30 Mhz.On Schematic T2.