Impedance between each END of a pot is constant. This is the load the phono cct "sees". It's the impedance from the WIPER to either end of the pot that varies.
Ladder-type attenuators can have different end-to-end impedances depending on the switch position.
As far as I remember, this is an issue that had raised a couple of times in various forums. What was Allen referring to was the ladder type stepped attenuator, which does not present the same input resistance in all settings of the switch. Allen suggested (and used in his products) series type stepped attenuator.
Note that for RTP3C, being a balanced design, you need one stereo attenuator for each channel - and this is how the VSE RTP3C is made. Since this seemed inconvenient to me (I mean, having to deal with two switches every time you want to adjust the volume), I chose to use instead a 4-gang stepped attenuator (and, for the same reason, a 4P6T input selector switch).
Evangelos
Hi Vincent,
My PSU looks like his:
An externally hosted image should be here but it was not working when we last tested it.
I checked that the 1.5 uF total capacitance before the choke is correct in this case - the value of 2uF, that I quoted earlier in the thread, was in my head because I have since built an active crossover using the same circuit that drew a little more current than the RTP3 and so needed slightly more capacitance.
The shunt reg pulls about 70 mA from the raw DC supply.
Alex
In my opinion you really have to use a stepped attenuator with the RTP3: the need for a 50K load for the phono stage means you can't use law-faking resistors on the pot (they would change the resistance of the pot), top-quality matched dual log pots are almost impossible to find, and linear pots have really poor control and matching at low volumes. I resorted to using 10K pots with a 40K series resistor on the input to get any real range at the lower end, but of course lost some of the upper end of the range as a result.
I have designed a 24-way stepped attenuator for my RTP3 to replace the Sfernice dual pots I have in there now. I have collected all the resistors, and am now waiting for the UK ELMA distributor to get hold of a couple of 24-way 4-pole switches for me.
I also think stepped attenuators are much better suited to dual controls than potentiometers, since you have repeatable volume settings.
Alex
Last edited:
The convenience is an issue. However, for best separation between the 2 channels, 2 separate volume controls and input switches are recommended.
Thanks, I assumed they were Elma or Shallco, I just havnt seen a constant impedance set up before. I guess an attenuator transformer would work too, more pricy though.
A transformer has an impedance that depends on the tap selected, so (as Allen pointed out a while ago) would be unsuitable in this case.
Alex
In fact, I had seen this circuit, but I thought you were referring to something else.
This is pretty much like the one published in the DPH1A manual that I have (see attached image), except that it uses 2SK170 FETs and there is no trimmer for adjusting the sinking current. Although, a thing that's not clear to me is that, in the ccs you cite, there is a connection between the drains of the FETs and the 1M resistor, while in mine the 1M resistor is grounded. Is yours in the RTP3C manual?
Attachments
My mistake - the 1M resistor is indeed grounded in my circuit.
The original also used 2SK170 devices: I changed the JFET type because the BF245C FETS were cheaper and easier to get hold of (Allen approved, as long as the gate stopper resistors were raised from 100R to to 1K, which I did). I added the trim pot to adjust the current - the original circuit assumed that the FETs were well enough matched not to need trimming.
Alex
Yes, this is a much better implementation of the ccs - it reminds me the one Allen used in the PP-2C - see my post here:
http://www.diyaudio.com/forums/tubes-valves/151733-vacuum-state-rtp3c-25.html#post2857037
It's far better than the original as it permits the user to balance current through the two legs of the differential stage. I will have to modify mine.
Thank you, Alex.
Evangelos