I've asked questions about this before about how to use it as a plate load. How would you use it as a current sink? Would you place a resistor above the "A" terminal and connect the "G" terminal to the top of the resistor?
http://www.ixys.com/98704.pdf
http://www.ixys.com/98704.pdf
No, don't use a resistor at the A. In fact you can use it as a sink just like the diagram in the PDF data. just connect the resistor (eg 100 Ohms for 25mA) from K to ground, connect G to ground, and connect the circuit you want to take a constant current FROM - to the A terminal.
Be sure to heatsink the device adequately to dissipate I(mA)*(Vdrop from A to K) in mW. Take care if you are experimenting with a current source in a valve cathode circuit, since if it goes open-circuit you will get a big voltage between the two points going to A and K & A and G.If you use figure 1, attach the Tubes cathode to A.
ground G and place the resistor K to Ground.
for 15 ma, the resistor appears to be about 200 ohms.
Basically looks a bit like a triode from a bias standpoint.
put negitive voltage on the K pin and ground the input, G.
AC is very like a pentode, BTW.
Important notes, indirectly heated triodes only. Directly heated triodes need a more complex treatment.
By the way, nice part. with 350v max, I bet its not cheap.
Doug
Using it as a plate load, you connect pin A to your B+ supply, and the bottom part where G and Rk meet, to the tube's anode. The amplified signal is also taken from this last point.
Note that the min dynamic resistnace is 10k, if you need more, it is better either to take a normal resistors or a real (discrete) current source. Also, the dynamic resistance can vary VERY wildly, so unless you have soem gain-stabilising feedback in your amp, it's gain will also vary all over the place.
Jan Didden
Note that the min dynamic resistnace is 10k, if you need more, it is better either to take a normal resistors or a real (discrete) current source. Also, the dynamic resistance can vary VERY wildly, so unless you have soem gain-stabilising feedback in your amp, it's gain will also vary all over the place.
Jan Didden
Hi G
It looks like they corrected the data sheet. It now shows a minimum value of 160K at ~10ma.
I have measured one of the 10M45S parts in my CCS test setup. It performed very well for a single part CCS. It's performance was almost 10X better than the single Supertex DN2540. A cascade pair of DN2540's is a whole different story with very good performance. Having only one of the 10M45S parts I haven't been able to try cascading them.
You can look at the graphs of CCS measurments made to date on my webpage. Look in the CCS measurments section.
Gary
Gary P's DIY page
It looks like they corrected the data sheet. It now shows a minimum value of 160K at ~10ma.
I have measured one of the 10M45S parts in my CCS test setup. It performed very well for a single part CCS. It's performance was almost 10X better than the single Supertex DN2540. A cascade pair of DN2540's is a whole different story with very good performance. Having only one of the 10M45S parts I haven't been able to try cascading them.
You can look at the graphs of CCS measurments made to date on my webpage. Look in the CCS measurments section.
Gary
Gary P's DIY page
Hi TC,
I use a single 10M45S as a constant current source in front of the plate of the preamp or driver tube. The result is quite good. I haven't tried cascoding the 10M45S as I don't how to implement it.
I think choky's idea is worth trying as you will have better regulations (both current and voltage) to the tube.
I use a single 10M45S as a constant current source in front of the plate of the preamp or driver tube. The result is quite good. I haven't tried cascoding the 10M45S as I don't how to implement it.
I think choky's idea is worth trying as you will have better regulations (both current and voltage) to the tube.
Hi Choky & DWong,
Thank you for your information. I think using 10M45s in series with the 0D3 is a good idea. I will try.
I will also use the 10M45s as the plate load of the 5676 gain stage. But just want to know how can the 10M45s be cascaded to improve their performance as the plate load.
Regards,
T.C. MA
Thank you for your information. I think using 10M45s in series with the 0D3 is a good idea. I will try.
I will also use the 10M45s as the plate load of the 5676 gain stage. But just want to know how can the 10M45s be cascaded to improve their performance as the plate load.
Regards,
T.C. MA
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.