Its a competent enough meter nonetheless, including the useful temperature function and its probe. Your Thorens TD321 is a capable classic turntable by the way. Good construction quality too, it was spot on for speed after all these years.
Salas, others,
A wild and quayzee one in the morning. I was thinking of the concept of parallelling input devices, in this case the 6N2PEV. They are farly easy to get and one could match up a bunch quite easily.
Morgan Jones eliminates the 12AX7 high mu tube types in this service due to noice and high grid capacistance ( I think he refers to MM in the later case) in Valve Amplifiers 4th Ed, but he also talks himself warm over the subject of parallelling input devices.
The pro's I can think of is higher gm, lower ra and extra current to drive the eq network. Every doubling would double the positive effects.
Of course one would need to recalculate the network (R5 I think is enough?). Maybe the Itch could serve as a hq MM riaa, and the expensive part, the SUT could be dropped?
Was I sleepless in vain?
A wild and quayzee one in the morning. I was thinking of the concept of parallelling input devices, in this case the 6N2PEV. They are farly easy to get and one could match up a bunch quite easily.
Morgan Jones eliminates the 12AX7 high mu tube types in this service due to noice and high grid capacistance ( I think he refers to MM in the later case) in Valve Amplifiers 4th Ed, but he also talks himself warm over the subject of parallelling input devices.
The pro's I can think of is higher gm, lower ra and extra current to drive the eq network. Every doubling would double the positive effects.
Of course one would need to recalculate the network (R5 I think is enough?). Maybe the Itch could serve as a hq MM riaa, and the expensive part, the SUT could be dropped?
Was I sleepless in vain?
It would take far too many. 10 to reach 21mS gm. Oscillations and Miller wait around the corner. Maybe 6C45P up to 4 max to start thinking low noise mc direct to the grid. Second stage would need a more gainy tube to cover for the lower mu input type. 60dB final gain sans SUT under 1nVrtHZ at least is a push. Careful there not to lose the input stage noise domination. Maybe move the RIAA on output and add a cathode follower or a source follower MOSFET.
Ok thanks Salas. I should have reckoned that it would have been unlikely for me to dream up a working scenario.
Lets say we implement the concept in the current set up, with MC SUT and two stage design, same tubes. Do you think it would be beneficial in terms of noice and drive current? As far as i grasp it R5 would be more in control on the total series resistance also.
Or is it likely that it just creates other mentioned problems?
Lets say we implement the concept in the current set up, with MC SUT and two stage design, same tubes. Do you think it would be beneficial in terms of noice and drive current? As far as i grasp it R5 would be more in control on the total series resistance also.
Or is it likely that it just creates other mentioned problems?
Last edited:
Maybe a double up would not hurt but it has to be calculated/simulated with a likely SUT source impedance to make sure about the doubled Miller influence. Balanced primary SUT is your hum killer friend, not to be easily dismissed for his noble services. Gives better 1/F field to the tubes also, they are not that happy there with low source impedances like MC carts have, even when their 1kHz noise figure looks nice.
I still havent choosen SUT yet, nor new PUP. Yet to many projects on the bench, one of them yours NJFET RIAA. Jensen doesnt answer my emails about selling bare SUTs so I guess I am leaning to something of Pers gear up in Norrtälje.
By the way, can you hint me how to calculate total series resistance in the riaa network in SRPP setup? Cant figure out what to divide the lower tubes Ra with.
By the way, can you hint me how to calculate total series resistance in the riaa network in SRPP setup? Cant figure out what to divide the lower tubes Ra with.
Its not srpp its mu follower we use in this phono. So the source impedance on the upper output where we connect the eq network will be 1/gm or ra/mu. You would want to know the effect of the Miller on the secondary from the SUT manufacturer. Some have an optimum or a certain recommended range for capacitive load.
Mu follower, thanks for the correction. Aha ok cool thank you. That meens that in this config the relative high ra of 6N2PEV is not so influencing? And in that case only the extra current would benefit from parallelled tubes?
Cant find any info on prefered capacitive loading in Lundahls documentation. Maybe I should write him and ask.
Cant find any info on prefered capacitive loading in Lundahls documentation. Maybe I should write him and ask.
Found a thread where Kevin Carter recommends www.lundahl.se/pdf/1550.pdf for 10x5. 6dB gain
Hi,
@ Stajo: If you're going to be using a HOMC then you probably won't even be needing the SUT.
You may even be overloading the first active stage with such a cartridge.
I assume Salas has designed the gain for MC cartridges and that the preamp is intended to be hooked up to a line level preamp doing the volume control and source selection?
Problem is that if you discard the SUT then the MC can't be loaded since the first stage is using the 47K resistor for biasing the lower triode. No big deal but it will require some redesigning.
OTOH, if you want to use an tubed headamp instead of the SUT then that can be done but it was designed for lower output MCs. IOW, the HOMC will likely overload that one as well.
Either way, I think Salas can advise you better than I can on this.
Ciao,
@ Stajo: If you're going to be using a HOMC then you probably won't even be needing the SUT.
You may even be overloading the first active stage with such a cartridge.
I assume Salas has designed the gain for MC cartridges and that the preamp is intended to be hooked up to a line level preamp doing the volume control and source selection?
Problem is that if you discard the SUT then the MC can't be loaded since the first stage is using the 47K resistor for biasing the lower triode. No big deal but it will require some redesigning.
OTOH, if you want to use an tubed headamp instead of the SUT then that can be done but it was designed for lower output MCs. IOW, the HOMC will likely overload that one as well.
Either way, I think Salas can advise you better than I can on this.
Ciao,
Frank, ty, yes. Well I woke up at 4 this morning and as any sound citizen I stared at the ceiling spinning of the idea of using high mu tubes in parallell and by so reducing their downsides. It seemed better then, then it does now.
Then one thing lead to the other. Hornets nest brain...
Well I have a bunch of D3a's rounded up for that purpouse here somewhere. I was thinking to go balanced even, we'll see.
But... 6dB is 1:2 no? 2,5 mV times 2 is 5 mV. Should work in this config, or?
Then one thing lead to the other. Hornets nest brain...
Well I have a bunch of D3a's rounded up for that purpouse here somewhere. I was thinking to go balanced even, we'll see.
But... 6dB is 1:2 no? 2,5 mV times 2 is 5 mV. Should work in this config, or?
Using 1N4148 or IR LED for cathode bias the leak won't develop much across the 47K. There was a user in this thread with 10X4 direct on grid and very satisfied BTW. 45-48dB gain is suiting it fine. 10X5 its 150 Ohm cart (better for 1/F and the tube than normal MC) also higher than 2.5mV at 5cm/sec. SUT is certainly not needed for it and the Itch. That user was loading it lower than 47K around 3.3-5K.