When driving a capacitive load, transformers have two main resonances:
1) HFres = series resonance between capacitive load and leakage inductance resulting in impedance minimum.
2) LFres = parallel resonance between capacitive load and primary inductance resulting in impedance maximum.
In either case, the resonance frequency is proportional to 1/sqrt(LC).
So, if C is increased by a factor of 2 with L remaining constant, resonance frequency decreases by sqrt(2) = 1.414.
Note that the total C used for this calculation would be the sum of the transformer winding capacitance and the externally connected capacitive load.
In the case of your measurements for capacitive loads of 300pF & 460pF, it would be expected that the resonance frequencies would decrease by 10% - 20% depending on the transformer winding capacitance. However your measurements show decreases of 60% - 100%.
Any thoughts on what is causing this discrepancy between theory and measurement?
Attachments
Wait a minute.....
if those curves are developing peaks as a result of changes in load (as well as quite peaked input-impedance curves and marked resonances right in the treble region), how can the panel output be flat? Hard to say what load our panels are presenting isn't it? Aren't we looking for flat (or at least, linear) voltage to the ESLs?
Ben
if those curves are developing peaks as a result of changes in load (as well as quite peaked input-impedance curves and marked resonances right in the treble region), how can the panel output be flat? Hard to say what load our panels are presenting isn't it? Aren't we looking for flat (or at least, linear) voltage to the ESLs?
Ben
Last edited:
In general yes, we are looking for linear voltage output(ie frequency independent amplitude) at the secondary of the step-up transformer. The peaks and dips in the impedance won't affect the output voltage if the primary is driven by a low impedance source, which fortunately most amplifiers are. However, the peaks and dips in the impedance have a direct affect on the current drawn from the amplifier at different frequencies when it is producing the linear voltage response. Frequencies where the impedance is low require more current from the amplifier than frequencies where the impedance is high.
Maybe my amplifier is one of the reasons. It has a -3dB point of 70kHz so the high frequencies are not very reliable.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.