wanna deep bass from small scale amp - build one with pure Square Law OS, meaning - no source resistors
you can cheat, for test - put 10mF across each existing source resistor, on appropriate amp
you can cheat, for test - put 10mF across each existing source resistor, on appropriate amp
Hi ZM, what about the F5 Turbo parallel diodes "trick"? - Wouldn't that work well on an F6 also?
diodes across source resistor is not the same as cap across or totally without source resistors
I believe that function of diodes is explained in F5T article, but - in short - when current through source resistor result in voltage same as diode voltage threshold, diode is taking all additional current through, without further limiting
cap across source resistor is sorta cheating - faking that resistor isn't there at all, in AC domain
I believe that function of diodes is explained in F5T article, but - in short - when current through source resistor result in voltage same as diode voltage threshold, diode is taking all additional current through, without further limiting
cap across source resistor is sorta cheating - faking that resistor isn't there at all, in AC domain
Yes, the big cap across resistor sounds like a great idea! In fact, I wonder why I havn't seen it before, and why it's not used more.
- Too good to be true? 🙂
For a big hi-voltage amp, the price and size of these caps would be problematic, but for F5 or F6 with just a single pair of outputs, a 10mF 35v cap would be simple enough to do.
But the diode trick is also neat, and seems quite effective.
So everyone doesn't have to look up the PDF, here is a quote from https://www.firstwatt.com/wp-content/uploads/2023/12/art_f5_turbo.pdf :
"...MUR3020W, rated at 400A peak and having an equivalent resistance of about .03 ohms per parallel pair.
Inserting these in parallel with the 0.5 ohm Source resistance gives us an impedance which is very low at high currents but devolves to the 0.5 ohm value at less than an amp."
- Too good to be true? 🙂
For a big hi-voltage amp, the price and size of these caps would be problematic, but for F5 or F6 with just a single pair of outputs, a 10mF 35v cap would be simple enough to do.
But the diode trick is also neat, and seems quite effective.
So everyone doesn't have to look up the PDF, here is a quote from https://www.firstwatt.com/wp-content/uploads/2023/12/art_f5_turbo.pdf :
"...MUR3020W, rated at 400A peak and having an equivalent resistance of about .03 ohms per parallel pair.
Inserting these in parallel with the 0.5 ohm Source resistance gives us an impedance which is very low at high currents but devolves to the 0.5 ohm value at less than an amp."
Papa used them ( as reminder to all of us) in one of F5-like small amps
lemme see in files ........
frifritfrit (sound of ZM rummaging through FW folder)
yup, 2021, DIY ACA Mini
go here: https://www.firstwatt.com/articles-and-reviews/
SuperCap across upper mosfet source resistor, switchable
As Ben wrote - in Toob world done all the time ......... as way of excluding local degeneration
Though, much easier, counting resistor values involved
Papa is Devil of Yore
lemme see in files ........
frifritfrit (sound of ZM rummaging through FW folder)
yup, 2021, DIY ACA Mini
go here: https://www.firstwatt.com/articles-and-reviews/
SuperCap across upper mosfet source resistor, switchable
As Ben wrote - in Toob world done all the time ......... as way of excluding local degeneration
Though, much easier, counting resistor values involved
Papa is Devil of Yore
Ahh right, as the cap doesn't see full rail voltage. That makes it much simpler.
I remember seeing the Supercap described for the ACA Mini, but it's in series with a 1r, so that is kind of a different usecase - to alter the harmonics.
Edit: Forgot to paste in the snip..
I remember seeing the Supercap described for the ACA Mini, but it's in series with a 1r, so that is kind of a different usecase - to alter the harmonics.
Edit: Forgot to paste in the snip..
Last edited:
Wasn't the point of the diodes in the F5 turbo v2 simply to give more power (or amperes), by reducing resistor losses? At least that is what I get from the description 🙂
-But perhaps we are getting a bit OT here. So perhaps we can bring it back by asking: Could eg. the F6 diamond buffer (Tungsten/MikeRodrig27 version) be improved by adding parallel Supercaps to the source resistors? - or would there be some drawbacks?
-But perhaps we are getting a bit OT here. So perhaps we can bring it back by asking: Could eg. the F6 diamond buffer (Tungsten/MikeRodrig27 version) be improved by adding parallel Supercaps to the source resistors? - or would there be some drawbacks?
Last edited:
These?
https://www.parts-express.com/Molex...MItOGe38q2iQMVDxmtBh0zbxlMEAQYAiABEgKoAfD_BwE
Do they introduce resistance into the circuit? Can they handle the heat? How long are the flying leads before (if) they introduce noise in the circuit? It is an interesting idea for sure. From a packaging standpoint it's brilliant.
Yes, great looking amp.
“If I did Thing X would the results be pleasing?”
One way to learn the answer is evident.
One way to learn the answer is evident.
The link is not the MiniFit Jr which is capable of 9A. For Class A amps with typical bias currents of 2A to 3A, they work well. They don’t introduce noise and are reliable. Keep flying leads as short as possible. Use silicone jacket high strand count RC drone battery wire.
Here is pdf of various MiniFit jr
https://www.mouser.com/datasheet/2/276/050361766_sd-1129127.pdf
For MOSFETs, I mount the MOSFET to a snubber board then solder wire to snubber board (which doubles as clamp washer). Use connector at the PCB where amp is.
The ones you linked are rated even higher current.
Faston spade 0.250in are good for 20A.