"A mistake is always forgivable, rarely excusable and always unacceptable. " (Robert Fripp)
View all posts by Ale Moglia
4 thoughts on “Testing Robustiano”
Hi Ale,
Quite a progress you did…
Following your work on the Robustiano and love the idea of solid state help with, hopefully with no degradation of sound and much cheaper compare to valves.
How is the sound compare to triode mode and filament bias?
At this point I am struggeling with the need for more power (class A sound), but the only way in triode mode is to use many of them in parallel ( for my desire of 10W I would need a sextet and 210mA…ouch). I tried class A2, but not liked as much as A1 despite a pretty smooth transition from A1 to A2.
Love the picture….so many intertwined wires..
Best,
Radu
Hi Radu,
Thank you, this is a very enriching experience indeed.
It is hard to answer the question and will be unfair to this circuit to attempt doing so given the testing conditions I put this under. The power supply has a voltage feedback-type regulator and never sounded great. The filaments are also powered up by a simple DC supply, no Rod Coleman regulators here, however, this is very quiet. What is even worse, the driver is a bare-bone Monacor SX200 full range that hasn’t had a proper burn-in process.
I did some minor tweaks by changing the MPSA42 driver bypass capacitor for a proper WIMA MKS2 10uF/63V one. Also added an additional 30uF cathode bypass oil cap to the existing 30uF in ultra-path connection. Played for some time with the feedback pot and couldn’t notice a significant difference between 38-45mA of bias anode current. I was expecting to notice this triode-like to more of a pentode-like spectrum, but that wasn’t evident to me. Perhaps is due to testing conditions, however distortion levels are low so not sure what might be.
Regarding your PSE idea, I think 8W should be plenty enough, but not sure what the sensitivity of your speakers is. You can consider fixed-bias, that should work very nicely for 4P1L. Also perhaps biasing the 4P1L at 300V and exceeding the anode voltage specs. It should work perfectly fine at 300V and this will give you some further room to extract more power. For example a pair of 4P1L with an Zaa of 4K can give about 5W (THD 3.42%) in A1 when biased at 300V/60mA.
Driving these pairs, triplets or quartet will require a stiff cathode follower to get the most of them when handling the input capacitance at HF. This may have been one of the reasons why you didn’t like A2. A2 is hard to implement effectively and a normal driver (e.g. IT coupled) is not the best choice. In fact, the A1 to A2 transition is the trickiest thing!
Ale,
I am still using Markaudio Alpair12P in a small reflex cabinet( not ideal and for that reason I will built some more adequate cabinets this winter or when have more time). These are 92dB sensitivity at 1W/1m. I am very happy actualy with my PSE 4P1L when listening at about 6 feet. My living room is medium size and I need to be able to listen from about 12 feet and need 2W now. When listening to classic piano, lets say Chopin, the dynamic range could easily be 10dB….oh well, now I need a 20W amp to satisfy my listenning needs…
I guess the answer is not simple and probably I need to change both speakers and amp power in order to solve the dilllema. I listened to a Tron amp with 211 and Bastianich 100dB speakers and what a difference…tried with my 3W PSE on those. Bastianich and didn’t like the sound( the amp had hard driving the speakers).
I tried a quad 4p1l with fixed bias ( at -23V using Rod’s regulator on a 65ohm load..the way he recommended) driven by 6E6p-ev through a LL1671 on 4.5:1 connection. The sound was ok and measured well ( I was able to reach 12W before hard clipping) but not close to the sound of my reference amp PSE with Etude1.
Perhaps you’re right about driving in A2 with IT. And I didn’t like very much the transition to A2. Never tried a cathode follower to see the difference.
I have stay in A1 with 4P1L….just to many 4P1L’s to reach a decent power level.
That’s why I am so excited to see your progress.
When you said 300V, did you tried it before ( the highest I went was 270V) ? Is the 8 w for a PSE pentode or triode? And also I like to stay under 2% THD where I cannot hear it.
Thanks Ale and good luck with your experiments.
Best,
Radu
You should aim for a clean 5W in A1 then. I’d strongly suggest to implement a cathode follower to drive the 4P1L PSE pair/triplets/quartets. You will notice a huge difference in the treble as the input capacitance is effectively charged/discharged by the follower. Easy to implement in your setup with a DN2540 follower. You could also use it to set the bias in a fixed bias circuit and by doing so you can move the coupling cap to the follower input and therefore making it rather small and then coupling it in DC to the grid.
I haven’t personally tried the 4P1L at 300V, but should withstand that for sure. It’s a very sturdy little fellow this Russian valve and should work that way for sure in triode-mode.
Agree with the THD level, I just shared maximum drive for a pair in this mode. If you want more power, then you could look at transmitter valves, but that is another story.
Hi Ale,
Quite a progress you did…
Following your work on the Robustiano and love the idea of solid state help with, hopefully with no degradation of sound and much cheaper compare to valves.
How is the sound compare to triode mode and filament bias?
At this point I am struggeling with the need for more power (class A sound), but the only way in triode mode is to use many of them in parallel ( for my desire of 10W I would need a sextet and 210mA…ouch). I tried class A2, but not liked as much as A1 despite a pretty smooth transition from A1 to A2.
Love the picture….so many intertwined wires..
Best,
Radu
Hi Radu,
Thank you, this is a very enriching experience indeed.
It is hard to answer the question and will be unfair to this circuit to attempt doing so given the testing conditions I put this under. The power supply has a voltage feedback-type regulator and never sounded great. The filaments are also powered up by a simple DC supply, no Rod Coleman regulators here, however, this is very quiet. What is even worse, the driver is a bare-bone Monacor SX200 full range that hasn’t had a proper burn-in process.
I did some minor tweaks by changing the MPSA42 driver bypass capacitor for a proper WIMA MKS2 10uF/63V one. Also added an additional 30uF cathode bypass oil cap to the existing 30uF in ultra-path connection. Played for some time with the feedback pot and couldn’t notice a significant difference between 38-45mA of bias anode current. I was expecting to notice this triode-like to more of a pentode-like spectrum, but that wasn’t evident to me. Perhaps is due to testing conditions, however distortion levels are low so not sure what might be.
Regarding your PSE idea, I think 8W should be plenty enough, but not sure what the sensitivity of your speakers is. You can consider fixed-bias, that should work very nicely for 4P1L. Also perhaps biasing the 4P1L at 300V and exceeding the anode voltage specs. It should work perfectly fine at 300V and this will give you some further room to extract more power. For example a pair of 4P1L with an Zaa of 4K can give about 5W (THD 3.42%) in A1 when biased at 300V/60mA.
Driving these pairs, triplets or quartet will require a stiff cathode follower to get the most of them when handling the input capacitance at HF. This may have been one of the reasons why you didn’t like A2. A2 is hard to implement effectively and a normal driver (e.g. IT coupled) is not the best choice. In fact, the A1 to A2 transition is the trickiest thing!
Cheers,
Ale
Ale,
I am still using Markaudio Alpair12P in a small reflex cabinet( not ideal and for that reason I will built some more adequate cabinets this winter or when have more time). These are 92dB sensitivity at 1W/1m. I am very happy actualy with my PSE 4P1L when listening at about 6 feet. My living room is medium size and I need to be able to listen from about 12 feet and need 2W now. When listening to classic piano, lets say Chopin, the dynamic range could easily be 10dB….oh well, now I need a 20W amp to satisfy my listenning needs…
I guess the answer is not simple and probably I need to change both speakers and amp power in order to solve the dilllema. I listened to a Tron amp with 211 and Bastianich 100dB speakers and what a difference…tried with my 3W PSE on those. Bastianich and didn’t like the sound( the amp had hard driving the speakers).
I tried a quad 4p1l with fixed bias ( at -23V using Rod’s regulator on a 65ohm load..the way he recommended) driven by 6E6p-ev through a LL1671 on 4.5:1 connection. The sound was ok and measured well ( I was able to reach 12W before hard clipping) but not close to the sound of my reference amp PSE with Etude1.
Perhaps you’re right about driving in A2 with IT. And I didn’t like very much the transition to A2. Never tried a cathode follower to see the difference.
I have stay in A1 with 4P1L….just to many 4P1L’s to reach a decent power level.
That’s why I am so excited to see your progress.
When you said 300V, did you tried it before ( the highest I went was 270V) ? Is the 8 w for a PSE pentode or triode? And also I like to stay under 2% THD where I cannot hear it.
Thanks Ale and good luck with your experiments.
Best,
Radu
You should aim for a clean 5W in A1 then. I’d strongly suggest to implement a cathode follower to drive the 4P1L PSE pair/triplets/quartets. You will notice a huge difference in the treble as the input capacitance is effectively charged/discharged by the follower. Easy to implement in your setup with a DN2540 follower. You could also use it to set the bias in a fixed bias circuit and by doing so you can move the coupling cap to the follower input and therefore making it rather small and then coupling it in DC to the grid.
I haven’t personally tried the 4P1L at 300V, but should withstand that for sure. It’s a very sturdy little fellow this Russian valve and should work that way for sure in triode-mode.
Agree with the THD level, I just shared maximum drive for a pair in this mode. If you want more power, then you could look at transmitter valves, but that is another story.
Cheers
Ale