I tried the below configuration and the results are very promising. Sound is as good as previously implemented SiC cathode bias.
Now looking at the TH and frequency response:
The stage performs as expected. I built some prototype PCB for the SiC filament bias. One board per channel. If there is any interest, I will run a bigger batch
Sounds great, as always.
Cheers
Ale
Which SiC diode did you use?
http://www.bartola.co.uk/valves/2015/05/09/sic-cathode-bias/
I would certainly be interested in some pcb’s for sic bias.
I ran a small prototype batch and made some errors on the PCB. Had to correct them. If there is enough interest will run it again.
What’s the PCB price ?
Hi Ale,
Nice find that filament SiC bias.
I’m also interested in a pair of those PCBs.
Regards,
Danny
Hi Ale,
Didn’t you mean “as good as filament bias” in your last post? I have the impression that you’ve been using filament bias a lot in your preamp incarnation, and not so much regular cathode bias.
Or did I miss something?
Best regards,
M.
Hi Maxime, I meant as good as the SiC cathode bias version I recently tried. I just updated the post. The fact that more current is going through the SiC diode, the dynamic impedance of the diode goes lower than 1 ohm which is great and also the diode is operated in the most linear region. As you can see on the THD plot above, I have no indication of additional distortion introduced by the diode. The array of 6 is perfect. You can either use 5 diodes should you wish to reduce the bias a tad. The diode in TO220 plastic can work without heatsink at this current. I’m also using same arrangement for the “Ba” triode which has 500mA of filament current and 500mW are dissipated on each diode. It works fine.
thanks for the question, It helped me refining the blog post!
Hi Ale,
did you also try the 4P1L with SIC filament bias ?
Cheers
Peter
Hi Peter, no. I won’t. 2P29L is a better choice. Still haven’t replaced the filament resistor for the SiC diodes.
I have ran a GB for Gyrator Board before to build 01a preamp. the result is wonderful.
Just wonder if any luck to try the new diode filament bias in HK
If you find enough folk out there in HK interested in the board…
Hi Ale,
I am interested in a couple of the diode board probably more later.
kind regards
Barry
Before anyone else keeps asking same question, I made a small batch of PCBs. The PCB holds 6 SiC diodes for up to 6V bias arrangement. It has a jumper for each diode so you can select different voltages if you want to reuse the board and experiment. Otherwise can be hardwired.
The boards will go on a first come first serve basis. Will post again when the boards arrive.
Which SiC diode did you use?
Ale,
Does the thd spectrum look the same between the two flavors of 01a preamp with SiC bias? I would think that the version of the 01a preamp running with only 4ma through the SiC diodes would have higher dynamic resistance and thus slightly lower distortion. Do the measurements show that effect?
Thanks for sharing the very interesting experimental results.
—Gary
Hi Gary,
What I have experienced is with the good 01a valves you should get about 10dB difference between H2 and H3. In this case, I get 20dB. However, I noticed that this value varies depending on the health of the DHT. These are very old valves! The above FFT proves that distortion isn’t higher due to the reduced dynamic cathode resistance. The degeneration produced by 25R filament resistor is minimum.
What I like seeing on the harmonic spectrum plot is the cadence of H2, H3, H4 and H5. Mind you that -77dBu is just 140uV!
Cheers, Ale
Hi Ale, I am very interested in the filament SiC bias but when I look at your schematic and voltages mentioned one thing I don’t understand. You are using a 5V voltage regulator as a voltage source but at the Cathode I see a voltage of 5.7V. I think your voltage regulator must than be a bit more than 10V because there must be a voltage drop of 5V from your filament if i am correct. I was also thinking if you use a voltage regulator as a current source for example a LM317 and adjust it with a resistor at 200mA or maybe close to 245mA. Maybe you can explain the 5.7V and also what you think about my idea with the current source.
Ignore the drawing typo. The regulator is set to 10.7V manually by measuring the voltage drop across the 01a filament pins to adjust to 5V
Ale, Is that a typo on Schematic where c1 is 510nf, or is it a change due to Diode Bias?
Nope, that’s the Russian cap I had at hand 🙂
Hi Ale, in some incarnations of your 01a preamp you used starved filament current (200mA). Why with SiC filament bias use nominal 250 mA? Thanks Jan
Hi Jan, over time I moved out from Filament starvation to thicker top plate as a way of controlling microphonic noise. Both are good option and you need to see what works best for you. This has nothing to do with the type of bias arrangement (eg SiC or resistor)