PRR said:> What is the input and output impedance of the filter?
What is rp is EF806s? The inductance of DV41 may matter. Parts 30 and 31 are in there also. Tube 42 is worked with NFB so approaches 200k input.
I basically just figured 120Hz with 5,000pFd to get an L. This may be 2:1 off, or more, either way. But it does mean a chunky part, not a tiny doughnut.
It's a 3rd order filter with different input and output impedances. The inductor should be about 150-200H.Murdock said:Thanks! But which formular did you use? If I use a second order cl high pass filter calculator with 5000pF and 120Hz I get about 350H...
What kind of filter is this?
abbey road d enfer said:It's a 3rd order filter with different input and output impedances. The inductor should be about 150-200H.
I cheated! I used a calc.Murdock said:Thanks for the answer!
Can you please elaborate on how you calculate that?
The output of the filter sees a Virual Earth stage with an input resistor of 200k. The input impedance of a VE-based stage is more or less that of the resistor that feeds it.How do you know the in and output impedance?
Actually, that's what I did in order to ensure I had not make a stupid power of 10 mistake, which is what happens frequently when scaling.I would like to simulate it in spice for different frequencies.
Indeed.150-200H seems rather high for just a lowcut filter.
That would increase the load presented to the preceding stage (increase load=decreasing the impedance); you would then run into headroom problems. You would also need to increase the output load of the filter, because with smaller L and larger C's, the damping factor would decrease seriously, resulting in a sharp hump at the turnover frequency. that would be easy but it would result in significant attenuation.Could i increase the two capacitors and decrease the inductor value for the same frequency?
There's a significant difference, the input and output Z's are identical; there are two 40k resistors feeding the filter, so they don't worry about loading the preceding stage, whatever the loss is, they compensate it in the subsequent stage.CJ said:here is a similar situation with the 76>
Except there's global NFB around; if there wasn't, what would be the use of the 2uF cap (marked 22)? That's why I penciled "less than 10k". Actually, it isn't significant; who said 120Hz is a magic number?PRR said:The source appears to be largely a triode plate resistance, goosed-up by unbypassed cathode resistor, with also 47k and a cap. I drew a line at 40k.
Correct; I didn't notice the 47k resistor. Duh!PRR said:> there's global NFB around
Oh, so you think if I give SPICE the wrong problem it will give a right answer to that wrong problem?
your sim directive should be in octaves not linear; type .ac oct 50 20 1kMurdock said:Wow, thanks alot for the informative answers!
I would like to simulate most of the circuit before I buy the parts.
Strangely I get very different results when I simulate in spice.
This is 24dB down at 100Hz...
Can you see what I missed in my circuit?
abbey road d enfer said:your sim directive should be in octaves not linear; type .ac oct 50 20 1k
Could you please post the whole schematic of the preamp?Murdock said:looking at the EAB MV63 schematic.
Enter your email address to join: