SSS Step filters - Here's what I think

[Matt J]

For those of you who have built these step filters, what brand type of capacitors have you used and which type do you think works best for this application?

- Matt J.

[martin manning]

Ryan C's SSS #002 layout and Erwin V's pcb marked up for modified filters.
[edit: note error in R Colgan layout shown in Red]

[Lothy]

Ryan C's layout and Erwin V's pcb marked up for the modified filters.

Thanks Martin.

[bcmatt]

Thanks for this Martin! I'm really impressed with this iteration of your updated filters. This looks to be the same as the schematic that I implemented yesterday to my SSS (The conversion from a YGL-3a). I had to search around to find your newest version. I can't right now find the original schematic (I think by Viperdoc) that I was using. (I'm swimming in every SSS schematic I can find right now, looking for options to try to tweak mine with).

Manning Filter Mod.png

I really like how even the increments are compared to what I had before. Each next increase in either highs or lows is what you would expect they should be and there are no weird surprising jumps, or too subtle of steps. Great job!

20240415_180451.jpg

I could only do 6 choices so a gave up the brightest choice and the first step in lows. I also had to parallel caps a couple times to get as close as possible to the right value with components I had on hand.

[martin manning]

Good to hear you are happy with the result! Yes I have made some small adjustments re the schematic at the top of this thread. What you posted just above is the latest. There are a couple of ways you could do six positions, one would be to recalculate all of the resistor values to keep the "taper," and make the 1 Meg total in six steps. The other would be to keep all of the original values and eliminate position 1 by using some convenient tie point in its place. I'd favor the second approach to maintain the step size.

[bcmatt]

Good to hear you are happy with the result! Yes I have made some small adjustments re the schematic at the top of this thread. What you posted just above is the latest. There are a couple of ways you could do six positions, one would be to recalculate all of the resistor values to keep the "taper," and make the 1 Meg total in six steps. The other would be to keep all of the original values and eliminate position 1 by using some convenient tie point in its place. I'd favor the second approach to maintain the step size.

I just realized I made a mistake implementing a proper elimination of the 1st step. I do need to add a 68k resister and move over the bottom leg of the .0068 cap to a tie in point. By omitting that resister, I wasn't using steps 2 through 7, but rather something more like 1.5 through 6.5 without the proper precision of the design since they no longer add up to 1M (only 930k).
I imagine the difference is subtle, but I may as well fix it since I already went through the trouble of rewiring it 95% of the way and I already like it so much more.

[martin manning]

To be strictly correct in eliminating position 1 you would do this:

[bcmatt]

To be strictly correct in eliminating position 1 you would do this:

OK, great, that's what I was thinking I would do.