Supertex dn2540 CCS Rset trimmer value

Alexontherocks · 3rd July 2012, 03:52 PM

I have read all available threads (also the "suggested" threads when writing this post) but to no avail.

I woul oike to use both the supertex dn2540 and IXYS 10m45s part as single device CCS

To do that I must select an appropriate Rset value (2.2k gate stoppers on each device).

The application requires 3mA and 5mA (both the supertex device and IXYS device will have to provide either one or the other). With a 2.2k gate stopper I thought of adding a 5k multiturn linear trimmer on the IXYS devices but I am not sure the same value would work for the supertex part.
The datasheet for the IXYS device does provide a graph with the Rset to current relationship however there the supertex datasheet does not.

Could someone please address the following questions?

1. Is a 5k multiturn pot a good choice to precisely achieve 3mA or 5mA from either device as a plate CCS? Is a larger value (ie 10k) better for such small currents?

2. What exactly is the correct value for each device (single IXYS and single SUPERTEX) to achieve 3mA and 5mA?

Thank you for any reply

Cheers

Michael Koster · 3rd July 2012, 04:45 PM

Quote:

Originally Posted by Alexontherocks

I have read all available threads (also the "suggested" threads when writing this post) but to no avail.

I woul oike to use both the supertex dn2540 and IXYS 10m45s part as single device CCS

To do that I must select an appropriate Rset value (2.2k gate stoppers on each device).

The application requires 3mA and 5mA (both the supertex device and IXYS device will have to provide either one or the other). With a 2.2k gate stopper I thought of adding a 5k multiturn linear trimmer on the IXYS devices but I am not sure the same value would work for the supertex part.
The datasheet for the IXYS device does provide a graph with the Rset to current relationship however there the supertex datasheet does not.

Could someone please address the following questions?

1. Is a 5k multiturn pot a good choice to precisely achieve 3mA or 5mA from either device as a plate CCS? Is a larger value (ie 10k) better for such small currents?

2. What exactly is the correct value for each device (single IXYS and single SUPERTEX) to achieve 3mA and 5mA?

Thank you for any reply

Cheers

All will be less than 1K. A smaller value pot is better to make the adjustment less sensitive. I'd use a 1K pot.

Alexontherocks · 3rd July 2012, 04:57 PM

I checked again the IXYS 10m45s datasheet. Actually according to the datasheet 1k would seem to be exactly the value needed for 3mA.

However the dn2540 part usually requires a higher value no?

Michael Koster · 3rd July 2012, 05:32 PM

Quote:

Originally Posted by Alexontherocks

I checked again the IXYS 10m45s datasheet. Actually according to the datasheet 1k would seem to be exactly the value needed for 3mA.

However the dn2540 part usually requires a higher value no?

I'm going by the datasheet and memory that the DN2540 cuts off below about 2.2 volts The official max Vgs cutoff is -3.5 volts which would require 3.5 mA through the 1K resistor. So I think you're pretty safe with 1K and the DN2540.

With the 10M45 I recall always needing a smaller resistance than the datasheet chart shows. Closer to the 10M90 spec actually...

Alexontherocks · 3rd July 2012, 09:07 PM

Thanks! I will start with 1k and see where that goes.

Fenris · 3rd July 2012, 10:32 PM

Those devices don't work well below 10ma. For values that low, I'd use a cascoded LM134/LM334 and 10m45s or dn2540. I use Walt Jung's circuit that he used for his article on constant current sources for the LM317 and 10M45s cascode, except the physical pinout of the LM134 is reversed from the pins of the LM317. The LM134/334 datasheet has values for the Rset that are more accurate and consistent than the depletion mode MOSFETS.

Alexontherocks · 3rd July 2012, 10:43 PM

very nice. I remembered the combo from the article...just a few questions.

1) Could you please define bad? Noise rejection/reliability...?

2) why replace lm317 with lm334? The schematic shown in the article seems to be "ideal" (words of the author) Is the load too low even for lm317+dn2540?

Thanks!

Alexontherocks · 3rd July 2012, 10:53 PM

Also...is the stabilization R/C circuit required even with lm334 operation?

Fenris · 3rd July 2012, 11:16 PM

The problem with the MOSFET alone or with the LM317 is that at low currents (less than 5-10ma) the dropout voltage becomes too low and they become either unstable or unable to keep regulation. The LM334 has a maximum of about 10ma while the LM317 has a minimum of around 5 ma. Below 5ma I use the LM334 with great success (see: New amp: 13DE7 SE hybrid). I use the stabilization network as cheap insurance - it costs little and can't hurt.

Alexontherocks · 4th July 2012, 05:33 AM

with a lm334 a good value trimpot would apparently be 100ohm. Correct?

Also in the schematic (by Jung) there is the R/C stabilization network but also a cap from B+ to ground. Could someone elaborate on that?

Thanks!
P.s. I read the thread you opened. Nice design. I for one like the mosfet in the output!

3rd July 2012, 03:52 PM	#1
Alexontherocks Account disabled at member's request Join Date: Jul 2009	Supertex dn2540 CCS Rset trimmer value I have read all available threads (also the "suggested" threads when writing this post) but to no avail. I woul oike to use both the supertex dn2540 and IXYS 10m45s part as single device CCS To do that I must select an appropriate Rset value (2.2k gate stoppers on each device). The application requires 3mA and 5mA (both the supertex device and IXYS device will have to provide either one or the other). With a 2.2k gate stopper I thought of adding a 5k multiturn linear trimmer on the IXYS devices but I am not sure the same value would work for the supertex part. The datasheet for the IXYS device does provide a graph with the Rset to current relationship however there the supertex datasheet does not. Could someone please address the following questions? 1. Is a 5k multiturn pot a good choice to precisely achieve 3mA or 5mA from either device as a plate CCS? Is a larger value (ie 10k) better for such small currents? 2. What exactly is the correct value for each device (single IXYS and single SUPERTEX) to achieve 3mA and 5mA? Thank you for any reply Cheers

3rd July 2012, 10:32 PM	#6
Fenris diyAudio Member Join Date: Jun 2004 Location: The last frontier	Those devices don't work well below 10ma. For values that low, I'd use a cascoded LM134/LM334 and 10m45s or dn2540. I use Walt Jung's circuit that he used for his article on constant current sources for the LM317 and 10M45s cascode, except the physical pinout of the LM134 is reversed from the pins of the LM317. The LM134/334 datasheet has values for the Rset that are more accurate and consistent than the depletion mode MOSFETS. __________________ Listen to the music through the stereo, not the stereo through the music.

3rd July 2012, 11:16 PM	#9
Fenris diyAudio Member Join Date: Jun 2004 Location: The last frontier	The problem with the MOSFET alone or with the LM317 is that at low currents (less than 5-10ma) the dropout voltage becomes too low and they become either unstable or unable to keep regulation. The LM334 has a maximum of about 10ma while the LM317 has a minimum of around 5 ma. Below 5ma I use the LM334 with great success (see: New amp: 13DE7 SE hybrid). I use the stabilization network as cheap insurance - it costs little and can't hurt. __________________ Listen to the music through the stereo, not the stereo through the music.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
DN2540 + LM317 cascode CCS?	rongon	Tubes / Valves	11	23rd December 2012 11:25 PM
DN2540 CCS questions	RollE2k	Tubes / Valves	6	19th March 2012 06:12 PM
Rset values for IXYS IXTP01N100D	JLH	Tubes / Valves	5	7th March 2012 06:24 AM
Supertex CL2	Mikael Abdellah	Parts	0	18th March 2010 01:05 PM
Help on Cascode DN2540 CCS	swordsman	Tubes / Valves	1	24th December 2008 04:48 AM


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

3rd July 2012, 04:57 PM	#3
Alexontherocks Account disabled at member's request Join Date: Jul 2009	I checked again the IXYS 10m45s datasheet. Actually according to the datasheet 1k would seem to be exactly the value needed for 3mA. However the dn2540 part usually requires a higher value no?

3rd July 2012, 09:07 PM	#5
Alexontherocks Account disabled at member's request Join Date: Jul 2009	Thanks! I will start with 1k and see where that goes.

3rd July 2012, 10:43 PM	#7
Alexontherocks Account disabled at member's request Join Date: Jul 2009	very nice. I remembered the combo from the article...just a few questions. 1) Could you please define bad? Noise rejection/reliability...? 2) why replace lm317 with lm334? The schematic shown in the article seems to be "ideal" (words of the author) Is the load too low even for lm317+dn2540? Thanks!

3rd July 2012, 10:53 PM	#8
Alexontherocks Account disabled at member's request Join Date: Jul 2009	Also...is the stabilization R/C circuit required even with lm334 operation?

4th July 2012, 05:33 AM	#10
Alexontherocks Account disabled at member's request Join Date: Jul 2009	with a lm334 a good value trimpot would apparently be 100ohm. Correct? Also in the schematic (by Jung) there is the R/C stabilization network but also a cap from B+ to ground. Could someone elaborate on that? Thanks! P.s. I read the thread you opened. Nice design. I for one like the mosfet in the output!

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read