Wire wound = inductance. It both resonated the termination and sense and was an antenna. If the reg has noise or oscillation, the phono's floor gets wild with high harmonic noise across the full band on display. If you try listen, that translates to weak signal and hum, buzz, & hiss, prominent above it. Its very obvious when there is an acting up reg. To get my levels you would need the same card and software, else its apples to oranges. -100dB @ 20Hz some weaker cards or algorithms show for your version, and usually can not resolve down to -140dB @ 20kHz that the Riaa filter shapes.They resolve to -130dB. If that's the case, its normal.
The VISHAY-RODERSTEIN MKP-1837 47 nf capacitor I ordered for the RIAA filter arrived as 33 nf.
While I wait for the 47 nf capacitors to arrive will someone tell me where this RIAA preamplifier fits into the chain? I was just reading Erno Borbely discussing the impedance of a attenuator above 25K ohms degrading the distortion performance of JFETS in general and 2SK170’s in particular. Does this RIAA preamplifier feed a line stage buffer? Does this RIAA preamplifier feed a line stage with gain? Where does the attenuator fall in the chain?
What happens if I take the RIAA output through my current 100K stepped attenuator into a tube buffer having 47K input impedance?
I want to use this RIAA preamplifier where the current tube RIAA is installed.
DT
All just for Fun!
While I wait for the 47 nf capacitors to arrive will someone tell me where this RIAA preamplifier fits into the chain? I was just reading Erno Borbely discussing the impedance of a attenuator above 25K ohms degrading the distortion performance of JFETS in general and 2SK170’s in particular. Does this RIAA preamplifier feed a line stage buffer? Does this RIAA preamplifier feed a line stage with gain? Where does the attenuator fall in the chain?
What happens if I take the RIAA output through my current 100K stepped attenuator into a tube buffer having 47K input impedance?
I want to use this RIAA preamplifier where the current tube RIAA is installed.
DT
All just for Fun!
EB was talking about a pot's value before a fet line pre and its filter effect when feeding a high Crss input fet's Miller (2SK170). The output Z of your phono is buffered and about 30 Ohm. Can drive with ease a line stage or a sensitive power amp. If you got 30dB gain or more in your valve amp and over 90dB sens speakers it may drive the system loud enough without a preamp even, or with just a unity gain pre.
Yes. Base noise is -98dB @20Hz drifting down to -132dB at 20kHz. That is a good number on my system and beats the V1.0.
Had to do the same base and gate stopper change to get it stable. 220R was not enough for the gate of the irfp9140. I did not move the compensation network on this channel however.
Listening to it now. Quite stunning the detail and soundstage. Instrument seperation is also greatly improved. Won't comment more until it passes my fatigue test, which it is doing very well with right now.
Salas, do you like the PHS software. I downloaded their demo but havn't installed it yet so that I may try it out when the second r-core arrives and I characterize this unit.
Had to do the same base and gate stopper change to get it stable. 220R was not enough for the gate of the irfp9140. I did not move the compensation network on this channel however.
Listening to it now. Quite stunning the detail and soundstage. Instrument seperation is also greatly improved. Won't comment more until it passes my fatigue test, which it is doing very well with right now.
Salas, do you like the PHS software. I downloaded their demo but havn't installed it yet so that I may try it out when the second r-core arrives and I characterize this unit.
ARTA is better for steady results and base noise with more cards IMHO. PHS is more professional interface.
So you got no problems after the initial debugging. Cool. Match the comp networks places and values at a point so to be typical. Is the Vout steady with heat rise in box? Normally the Jfet will be happier after a while for current in the Ref and maybe you would re adjust. When you will find their typical values for 42V in closed box after thermal rise due to heatsinks radiating, best thing to do is to replace the trimmers with good quality metal film resistors of proper value. Better quality and safest in the long run.
P.S. After it will play enough for your ears so you think you got the gist of it, because it offers the best resolution and extension, its better to make your rake angle, VTF, and C3 HF Riaa Cap final tunings with the 1.2 under the bonnet.
So you got no problems after the initial debugging. Cool. Match the comp networks places and values at a point so to be typical. Is the Vout steady with heat rise in box? Normally the Jfet will be happier after a while for current in the Ref and maybe you would re adjust. When you will find their typical values for 42V in closed box after thermal rise due to heatsinks radiating, best thing to do is to replace the trimmers with good quality metal film resistors of proper value. Better quality and safest in the long run.
P.S. After it will play enough for your ears so you think you got the gist of it, because it offers the best resolution and extension, its better to make your rake angle, VTF, and C3 HF Riaa Cap final tunings with the 1.2 under the bonnet.
Arta was nice and worked well, but the interface seems less intuitive as well the graphics output is less than PHS. PHS listed both my card and your usb soundcard as good. I plan on continuing with this hobby for a while and would like a good measurement system so I will consider both. Audio tester imho is too unstable and has questionable algorythms. It is also impossible to select bit depth and sample rate easily.
On the v1.2, I am debating on whether to make all new boards. The current ones have been worked over quite heavily and are a little sloppy from solder - desolder- solder-desolder etc..... Regarding the comp network I plan on returning it to the original schematic at some point after it has played for a while and I am sure that the modified version isn't required. Right now it does not appear to be needed and will be a PITA to squeeze it in if it is.
The voltage is holding quite well at 42.0+/- 0.1V in the box. Again I'll let it cook for a while and see where it drifts to. The box is pretty heavy and is a good heatsink so after about 3hrs, you can feel that it is slightly warmer than ambient but not in any sense warm to the touch. When I get the other r-core and build the PS chassis we will get a good steady state condition for drift as these r-cores are running around 55V filtered for single channel. Not sure if it is important to kick in a little R before the pi filter to drop the voltage to 52 given how cool the mosfets run.
P.S. Currently my sample bit depth is 32bits which tends to raise the noise floor vs. 16bits. It crashes on anything alse.
P.S.S. Will set everything up regarding VTF etc, once next project is completed. Rebuild of TT. Design is almost complete and am starting to experiment with DIY dampening mass composites. Will start a thread on this project if my material works as well as I hope.
On the v1.2, I am debating on whether to make all new boards. The current ones have been worked over quite heavily and are a little sloppy from solder - desolder- solder-desolder etc..... Regarding the comp network I plan on returning it to the original schematic at some point after it has played for a while and I am sure that the modified version isn't required. Right now it does not appear to be needed and will be a PITA to squeeze it in if it is.
The voltage is holding quite well at 42.0+/- 0.1V in the box. Again I'll let it cook for a while and see where it drifts to. The box is pretty heavy and is a good heatsink so after about 3hrs, you can feel that it is slightly warmer than ambient but not in any sense warm to the touch. When I get the other r-core and build the PS chassis we will get a good steady state condition for drift as these r-cores are running around 55V filtered for single channel. Not sure if it is important to kick in a little R before the pi filter to drop the voltage to 52 given how cool the mosfets run.
P.S. Currently my sample bit depth is 32bits which tends to raise the noise floor vs. 16bits. It crashes on anything alse.
P.S.S. Will set everything up regarding VTF etc, once next project is completed. Rebuild of TT. Design is almost complete and am starting to experiment with DIY dampening mass composites. Will start a thread on this project if my material works as well as I hope.
Last edited:
Best floor I get with 24bits in general.
Yes, that is what I meant, match to original comp places and values. It had nothing to do with OLG and Phase Margin after the yesterday experiments showed, it was layout lengths in the error amp interacting with its base, and triggering the heavier Crss 9140 as a combination as it proved. Keep your own layout's stopper values and be happy. Naah, I would not remake if it is OK electrically just because I used the Iron over it a few times. Has the battle scars and medal of honour now, don't make it back to a college boy.
Nice it stays put for voltage since the box is a good sink. Good luck with the TT stuff.
P.S. Any pics with the mini 1.2s @ work?
Yes, that is what I meant, match to original comp places and values. It had nothing to do with OLG and Phase Margin after the yesterday experiments showed, it was layout lengths in the error amp interacting with its base, and triggering the heavier Crss 9140 as a combination as it proved. Keep your own layout's stopper values and be happy. Naah, I would not remake if it is OK electrically just because I used the Iron over it a few times. Has the battle scars and medal of honour now, don't make it back to a college boy.
Nice it stays put for voltage since the box is a good sink. Good luck with the TT stuff.
P.S. Any pics with the mini 1.2s @ work?
Board looks battle worn, but victorious.
Here are a few quick photos. Not as exciting as it use to be. Still need to put the brick where the old torroid use to be. Otherwise there would be a lot of wasted space.
Ignore the plywood shelves in the stand. The glass I ordered hasn't arrived yet.
P.S. Audio tester crashes at 24bits as well.
Here are a few quick photos. Not as exciting as it use to be. Still need to put the brick where the old torroid use to be. Otherwise there would be a lot of wasted space.
Ignore the plywood shelves in the stand. The glass I ordered hasn't arrived yet.
P.S. Audio tester crashes at 24bits as well.
Attachments
Still need to put the brick where the old torroid use to be. Otherwise there would be a lot of wasted space.
I would put the Rcores and pre filter inside without screwing them in and see if they are acceptable before going de facto two box. But if its a brick in the end, paint it white & blue before you glue it in and stick a Mykonos postcard on it. Its better for the sound.
I like them. Very compact and looking like hiding ammunition in the bomb bay underneath.
24bit with Arta if you try maybe it works?
@Salas
This is my new njfet 2SK170BL stash for balanced phono RIAA:
Idss
8.16 x 2 pcs
8.26
8.33 x 2 pcs
8.34 x 4 pcs
8.38 x 2 pcs
8.39 x 2 pcs
8.45
8.49
8.51
8.52
8.53
8.54 x 2 pcs
8.71
8.80
8.88
8.98
Please let me know wich I will use & wich position 1st stage or 2nd stage (I need 16 matched pairs for balanced njft RIAA).
Thanks in advance
This is my new njfet 2SK170BL stash for balanced phono RIAA:
Idss
8.16 x 2 pcs
8.26
8.33 x 2 pcs
8.34 x 4 pcs
8.38 x 2 pcs
8.39 x 2 pcs
8.45
8.49
8.51
8.52
8.53
8.54 x 2 pcs
8.71
8.80
8.88
8.98
Please let me know wich I will use & wich position 1st stage or 2nd stage (I need 16 matched pairs for balanced njft RIAA).
Thanks in advance