My local electronics surplus store was open for business this week, without the silly 1 customer in store at a time limitation. They have had these LED heatsinks for a while, and I finally realized a good use for them.
The plan would be to use these as heat spreaders / extra dissipation for the hockey pucks, and attach the remaining TO-247 packages directly to the main channel sinks. The LED heatsinks have a nice smooth surface on the back side which will make a good thermal interface to the main sinks.
The plan would be to use these as heat spreaders / extra dissipation for the hockey pucks, and attach the remaining TO-247 packages directly to the main channel sinks. The LED heatsinks have a nice smooth surface on the back side which will make a good thermal interface to the main sinks.
Attachments
Interesting. Do you have a parts number?
(PS: The front side does not look very smooth, doesn't it matter?)
Last edited:
My ears say that the amp sounds different with different load values,
causing it to traverse a different load line, and we should be seeing this
on distortion analysis. A SIT-3 is a fine example, where the lowest
distortion is seen for the region roughly between 4 and 8 ohms.
You can listen for this with a given pair of speakers by putting different
values of resistance in parallel with the load.
OK. Spinning the magic A-pot to 0.89 is get something that looks pretty good by your criteria. I show the spectra for 4R and 8R loads, with peak output levels of 4V, 8V, 12V, 16V and 20V. The 2nd and 3rd harmonics are nearly identical, and in the fourier data the phase change of H2 between 4R and 8R loads is about 10 degrees at 4V increasing to 45 degrees at 20V. In the spectral plots, the the 4th and higher harmonic are affected by Rload values.
The nice thing about the A-pot is that its adjustment does not affect the bias current Vds of the SIT.
The nice thing about the A-pot is that its adjustment does not affect the bias current Vds of the SIT.
Attachments
I am still waiting for parts to arrive, but much of the chassis work is complete.
In the meantime, I did some work with the fourier .log data in the LTSpice simulations to display the distortion waveforms at different values for Vout and Rload. Below are the residual waveforms with the magical value of A=0.89, which corresponds to a Schade feedback ratio, R1/R2=A/(1-A), of about 8.09. They look good to me.
In the meantime, I did some work with the fourier .log data in the LTSpice simulations to display the distortion waveforms at different values for Vout and Rload. Below are the residual waveforms with the magical value of A=0.89, which corresponds to a Schade feedback ratio, R1/R2=A/(1-A), of about 8.09. They look good to me.
