Hmmm.. seems the results I posted earlier were a bit off 
. I couldn't help wondering why resonance was so low, I expected much better values so I dived into it a bit further
Well it turned out that the old lm3886 amplifier i hooked it up to for testing had a very large output inductor ruining all my results
So I pulled my good ol Leach from the rack and did some new testing:
This time I loaded the transformer with 450pf, primary directly connected to the leach. I measured output voltage using a 1:100 probe and determined the freq of the peak at the high end of the transfer curve:
- only one of the 5 wires in the first prim layer connected: 28 khz
- all wires in first prim layer connected: 37 khz
- all wires in first and second prim layer connected: 45 khz
Peak height was around 2db/6db/5db resp.
Adding 1 ohm in the primary circuit to dampen the peak gives a flat response and -3db@51 khz (both primaries connected). Not bad at all
Of course this coil is only half the final transformer, I'll have to add another coilformer on the same core to complete it.
That will halve prim dc-resistance, double sec resistance and double stepup to 1:100. Performance will probably drop a little because of the greater capacity seen from the primary in that case, depending on how well the amp handles this. Still I expect it to perfomr quite respectable.
For the next (final?) one I think I'll add a few turns in the primary as inductance is a bit on the low side now (7 mH). I intend to run it from 100 hz upwards, now looking at 4.5 ohms there. That would also mean adding one more layer in the secundary, to compensate the loss in stepup and probably even raise it a little. It may be beneficial as well to wire up the sec layers in a different way so the the hot end is in the middle of the winding, in stead of against the second prim layer (less influence of interwinding capacity).
Even better would be to find a core with higher permeability...
wire! more wire!
. I couldn't help wondering why resonance was so low, I expected much better values so I dived into it a bit further Well it turned out that the old lm3886 amplifier i hooked it up to for testing had a very large output inductor ruining all my results
So I pulled my good ol Leach from the rack and did some new testing:
This time I loaded the transformer with 450pf, primary directly connected to the leach. I measured output voltage using a 1:100 probe and determined the freq of the peak at the high end of the transfer curve:
- only one of the 5 wires in the first prim layer connected: 28 khz
- all wires in first prim layer connected: 37 khz
- all wires in first and second prim layer connected: 45 khz
Peak height was around 2db/6db/5db resp.
Adding 1 ohm in the primary circuit to dampen the peak gives a flat response and -3db@51 khz (both primaries connected). Not bad at all
Of course this coil is only half the final transformer, I'll have to add another coilformer on the same core to complete it.
That will halve prim dc-resistance, double sec resistance and double stepup to 1:100. Performance will probably drop a little because of the greater capacity seen from the primary in that case, depending on how well the amp handles this. Still I expect it to perfomr quite respectable.
For the next (final?) one I think I'll add a few turns in the primary as inductance is a bit on the low side now (7 mH). I intend to run it from 100 hz upwards, now looking at 4.5 ohms there. That would also mean adding one more layer in the secundary, to compensate the loss in stepup and probably even raise it a little. It may be beneficial as well to wire up the sec layers in a different way so the the hot end is in the middle of the winding, in stead of against the second prim layer (less influence of interwinding capacity).
Even better would be to find a core with higher permeability...
wire! more wire! 
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.