first audio test
I have now completed initial work on the electronics of the v2 prototype, and have given it its first audio test:
On my youtube channel (Industrial Baroque), you'll also find a number of other videos documenting the build process leading up to this point.
My initial impressions: the clean sound is good, and output is strong, seemingly exactly the same as the v1 prototype. However, each of the circuit sections beyond the basic clean signal path seem to need work. This is as expected: I knew some adjustments would be needed, and my plan has always been to get the basic audio system functioning, and then "dial in" the various circuit sections one by one.
Gain stages (boost and M-drive): activating either or both of these high-gain stages, and turning up the gain controls, soon leads to horrible high-frequency oscillations. I pretty much knew this would be the case, and I have a few ideas on how to mitigate this, now that the system is up and running. I have 1kV, 10pF capacitors to place across some or all of the 12AX7 plate-cathode gaps. I didn't start out with these already in place, even though I knew they were likely to be required, because I wanted to assess the extent of the issue and only add as many as are needed, in the problem locations.
Also, I'm planning to provide for tube shields on all the 12AX7s. The tube sockets I'm using, since they mount from the inside (which facilitates my construction technique of building and testing the entire audio system outside of the chassis, before final assembly), do not have provision for the attachment of tube shields. But I think I can cannibalize the bases of external-mount sockets, with the bayonet-type tube shields, and attach them from the outside when I mount the tube sockets from the inside.
The boost stage in particular is really just a "place holder" circuit, with no refinement, just a basic raw gain stage which I knew I'd have to make adjustments on after the fact. It really has too much gain, and is a major contributor to the squealing problem when activated. I will try putting resistor-ladder attenuators, at the input and possibly also the output. These would also be the place for capacitors to be added, to shape the frequency response, if the 10pF "degenerator" caps are not enough.
The v-mid EQ may need some work; I simply guessed at the R-C values to tune the frequency range, but I'll have to make adjustments by-ear to make sure it is operating in the most useful range.
The fx loop is a concern: it seems to have too little gain, much less than gain of 1. When activated with nothing plugged into the return jack, the signal passes through the send attenuator resistors with a gain of 1/20, and then to the return amplifier stages, which should have a gain of 100. So overall gain should be 5; I expected to start with that and then adjust the gain downward towards 1, by decreasing the send impedance and/or dropping the gain of one or both of the return stages. But instead, there is a precipitous drop in signal level when the loop is engaged. I don't yet know why this is happening. Maybe I have the wrong resistor values some place in the circuit, or maybe my assumptions about how my "opamp style" negative feedback gain stages should work, are faulty.
Anyway, I'm pleased to finally have tangible, audible results from this build, and my work going forward is clearly cut out for me.
Comments
Post a Comment