That should correlate, but the model with lumped elements does not describe all the details. I have often found that the practical results are slightly better than the simulation based on lumped elements, just like simulating a transmission line with not enough cells produces pessimistic results.The bigger cores are on back order so they want come soon
But from the 3k9 source measurement one can extrapolate the freq response of my toroid with a 220r source impedance, I know the high end behavior is more complex than a low pass filter, besides capacitance it has leakage inductance etc... it has self resonanse I cant see.
In the low end the -3db is at 2Hz because inductance rises at low Hz to 16 H.
At the high end the measurement indicates -3db roll off at 120KHz
Would it be correct to calculate self resonance freq. by having total capacitance from freq plot: 5n8 + measured leakage inductance?
Help Support GroupDIY:
I discovered an error in my earlier measurment. The source impedance wasnt 3k9 it was really 1K3. Because the 3k9 resistor was in parrallel with 2K...
So lets set the record straight. Its less flattering but.. High freq roll off -3db at 40K. And like Jakob predicted there was no increased inductance at lower freq. Inductance is flat....
That put the Lundahls -3db at 80k under the same circumstances.
The flaws of capacitance are magnified trough the transformers impedance convertion. My toroid and lundahl starts out with pretty much the same total capacitance. Lundahl has 1:25 impedande ratio and the toroid has 1:42 that pretty much accounts for the earlier hi freq roll off...
I measured leakage inductance across the primary: 50uH with shorted (secondarys)
The 1k3 is reflected to secondary as 55K, -3db at 7Khz = 413p. Total capacitance.
With a 50 procent bigger core plus 50 procent higher primary inductance 10H. I think my next IPT should rival the THD to the LL1538.
The big challange is to get low capacitance. Interwinding C will be lower with electrostatic sheild. I dont think it is possible to progressive winding by hand, but many small sektions in 20-30 turns range (by winding the first 10 turns side by side and the rest on top of each other) will lower C between turns and evenly distribute winding through out the core... The last thing is by lowering turns ratio to 1:5 hi freq response will almost double...
The goal is to rival the LL1538 so to even the plane field I will go for a 1:5 toroidal design next... and for a DIY design maybe that is the sane end point...
So lets set the record straight. Its less flattering but.. High freq roll off -3db at 40K. And like Jakob predicted there was no increased inductance at lower freq. Inductance is flat....
That put the Lundahls -3db at 80k under the same circumstances.
The flaws of capacitance are magnified trough the transformers impedance convertion. My toroid and lundahl starts out with pretty much the same total capacitance. Lundahl has 1:25 impedande ratio and the toroid has 1:42 that pretty much accounts for the earlier hi freq roll off...
I measured leakage inductance across the primary: 50uH with shorted (secondarys)
The 1k3 is reflected to secondary as 55K, -3db at 7Khz = 413p. Total capacitance.
With a 50 procent bigger core plus 50 procent higher primary inductance 10H. I think my next IPT should rival the THD to the LL1538.
The big challange is to get low capacitance. Interwinding C will be lower with electrostatic sheild. I dont think it is possible to progressive winding by hand, but many small sektions in 20-30 turns range (by winding the first 10 turns side by side and the rest on top of each other) will lower C between turns and evenly distribute winding through out the core... The last thing is by lowering turns ratio to 1:5 hi freq response will almost double...
The goal is to rival the LL1538 so to even the plane field I will go for a 1:5 toroidal design next... and for a DIY design maybe that is the sane end point...
Last edited:
True, but the capacitance to ground will increase. In most cases, the addition of an electrostatic shield increases the stray capacitance. And the leakage inductance will also increase. These are two factors that contribute to decrease the HF response.
So the main benefit of a electrostatic screen would be improved cmrr. Noise rejection is already good beacuse of symmetric primary. So maybe I should reject screen and focus on battle capacitance in the secondarys.... decreasing ratio and more symmetrically wound secondarys.... hmmm
Yes.
A xfmr primary is symmetric uo to a point, where the capacitances at both expremities of teh winding start to differe significantly, impairing HF CMRR.
Many input circuits, particularly German, used a trimmer capacitor to balance HF CMRR.
A particularly well balanced secondary will not result in a perfectly balanced circuit it it's connected to an unbalanced input, which is the most common case.
I wished Bill Whitlock (Mr CMRR here) would chime on it.
You sure your enemy is the distributed capacitance (winding-to-itself), not the interwinding capacitance, i.e. capacitance between pri and sec's?
Remember: distance is your friend here - have you tried physically spacing the windings a bit further by some extra layers of inter-layer-tape?
Coupling is usually excellent in toroids, you probably don't need the closeness of windings for efficiency..
/Jakob E.
Remember: distance is your friend here - have you tried physically spacing the windings a bit further by some extra layers of inter-layer-tape?
Coupling is usually excellent in toroids, you probably don't need the closeness of windings for efficiency..
/Jakob E.
DIY Arduiuno based Toroid coil winding Machine - Electric Diy Lab
In this post I'll show you how to make DIY Arduino based toroid coil winding machine at home, this machine have userinterface like LCD & rotary knob.
electricdiylab.com
He makes a nice job of it with a simple self contained controler and bread board approach .
Attachments
yes, but having the raw-wire spool inside the ring makes it usable only for relatively large core diameters
