Do you mean a 10k volume pot? The effective input capacitance of the 2SK82 under L'Amp operating conditions was a couple thousand pf as I recall.
Most of my samples have a microamp or so of leakage current.
I agree Telstar plus perhaps with HP's you do not need so many settings as with speakers in the volume control - perhaps only 6-7.
I need at least 24 steps, cutting away the outer zones (i.e. choosing resistors to provide fine control in the normal listening level).
ATM I'm choosing the switch to use, and i'll order next month.
ATM I'm choosing the switch to use, and i'll order next month.
I think you will spend more time soldering the attenuator than on the rest of the amp.😛
Good luck and keep us informed. What headamp will you compare it with ?
I have a PPA.
Good luck and keep us informed. What headamp will you compare it with ?
I have a PPA.
Hell yes! but i found one already mounted with Holcos... is my time soldering it worth 20£? i think yes.
For comparison:
-modded CK2III
-Marsh Headamp (to build right after L'Amp)
Same regulated psu (varies only voltage and transformer size, type is the excellent Nuvotem potted).
The SIT is the bigger of the lot, I think I'll order a 100VA transformer for mine (+35v/0.5A bias).
I'm playing with discrete designs so far, other 2-3 further down on the list (a JLH 69, an italian project called VHF-N and an F5 or aleph redux with laterals).
When I'm done with all the ones I think are worthy, I'll check opamps (not many output ones high current types avail).
I soldererd a lot of these good old ELMA audio attenuators.
Attachments
Yes! Elma are my first choice. Used by goldpoint and not hard to find in Europe.
But those i found are with single pcb (per channel) with little spacing, for smd resistors. I'll look further (or get the pre-mounted one)
But those i found are with single pcb (per channel) with little spacing, for smd resistors. I'll look further (or get the pre-mounted one)
Yours are stepped attenuators that have always a chain of resistors in the signal path. Mine only one resistor at each step. A couple of years ago I purchased the last two from a British distributor. They seem to be hard to find meanwhile and if you get some, they are quite expensive.
But have a look at these high grade attenuators from Poland: High Quality Audio & Industrial Attenuators and Passive Preamplifiers
Mine? I have got none yet. 😉
You mean a shunt attenuator, i plan something like that.
They had mixed quality issues, so I prefer to go with Elma.
I have read the comments at http://www.diyaudio.com/forums/parts/149049-khozmo-48-position-stepped-attenuators-6.html.
Sounds good, at least in my comprehension...
can i repeat my "newbie" question ? what is the effect of this input capacitance and how to cure it on the Sit ? 😀
The effective input capacitance and the output impedance of whatever drives it form a low-pass filter with cutoff = 2*Pi*R*C
So, driving the SIT with a 600 Ohm source gives:
6.28 * 600 * 2E-9 = 133 kHz
And, for example, hanging a linear 50k volume pot on the input, set at 50%, gives:
6.28 * (600+12500) * 2E-9 = 6 kHz (which is hi-fi for humpback whales)
So, driving the SIT with a 600 Ohm source gives:
6.28 * 600 * 2E-9 = 133 kHz
And, for example, hanging a linear 50k volume pot on the input, set at 50%, gives:
6.28 * (600+12500) * 2E-9 = 6 kHz (which is hi-fi for humpback whales)
Last edited:
thanks michael i understand , so the 10K and 47K resistor loading the gate is not in the impedance calcul ? or perhaps for high pass with the coupling capacitor ? i understand but slowly 
From Michael's calculation I get that a 1k pot would be ideal to not have HF rolloff.
My source is very low impedance so I could use a 1k pot, the problem being it unobtainium 😉
Therefore i'll build a stepped attenuator taking advantage of the 10k (which i erroneously called inpout resistance) in it.
My source is very low impedance so I could use a 1k pot, the problem being it unobtainium 😉
Therefore i'll build a stepped attenuator taking advantage of the 10k (which i erroneously called inpout resistance) in it.