Author Topic: DIY-SMU Project  (Read 74862 times)

0 Members and 1 Guest are viewing this topic.

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #100 on: December 04, 2021, 02:40:32 pm »
I am confused. the signal CLAMP/ is connected to an INPUT of U25 and yet is connected to the ORed outputs of LM393 U24? Furthermore I could not find the origin of the signal in the .ino?
I checked the latest versions of the schematics but this is still unchanged.
Dave, could you clear this up for me? Have you implemented this in soft already?
« Last Edit: December 04, 2021, 02:45:29 pm by RikV »
 

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #101 on: December 04, 2021, 07:31:58 pm »
Dave, is there any particular reason why you use Si8661AB-B (default output LOW) and Si8422AB-D (default output High)? Especially the AB-B seems a bit odd since the SS lines are active low.

ADG261... (5kV) will not fit on the PCB: WIDE SOIC. Use ADG161.. instead (3kV)
« Last Edit: December 10, 2021, 08:03:06 pm by RikV »
 

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #102 on: December 10, 2021, 12:11:13 am »
I just noticed this: on the main board schematic (V2) J2/pin4 is connected to net "GND". I am sure this should be net "-15V".
Dave, can you pls confirm this?
Also J2/1 is connected to GND, meaning Analog gnd is shifted to -nearly- +Out level. OK?

Just checked this on the board itself: It is OK!??? (Rev1) Has an error been introduced in V2 board?

PS: All these remarks are correct, however: they are covered in Dave's Blog. Shame on me! should have paid better attention. Sorry Dave.
« Last Edit: December 13, 2021, 11:50:24 pm by RikV »
 

Offline rpiloverbd

  • Regular Contributor
  • *
  • Posts: 157
  • Country: bd
Re: DIY-SMU Project
« Reply #103 on: December 11, 2021, 11:13:46 am »
Is it Microcontroller based, sir?
 

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #104 on: December 11, 2021, 04:53:18 pm »
Yes, it is. It has a Teensy3.2 and a smart Nextion display on board. It even speaks SCPI!
Read all about it here: http://www.djerickson.com/diy_smu/index.html
 

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #105 on: December 19, 2021, 07:52:12 pm »
Dave, could you please updat Github to the actual -incomplete- status?
 

Offline ducreux92

  • Contributor
  • Posts: 17
  • Country: fr
Re: DIY-SMU Project
« Reply #106 on: December 20, 2021, 04:07:36 pm »
Hi,
Why do not use the reference voltage of AD5686RBRUZ (2ppm/deg) instead of ADR431 voltage reference ?
 

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #107 on: December 21, 2021, 08:10:24 pm »
Ducreux, nothing keeps you from doing so, they are pin-compatible. I used a ADR4525CRZ. Maybe Dave used ADR431 because he had one hangin around? For reason of cost? Since the DAC is "only" 16 bit it does not matter much if it is 1 or 3ppm/K.
 

Offline ducreux92

  • Contributor
  • Posts: 17
  • Country: fr
Re: DIY-SMU Project
« Reply #108 on: December 22, 2021, 09:09:43 am »
Hi,
I agree with you but i think a single internal voltage reference (AD5686RBRUZ) for DAC 16 bits and ADC 24 bits is  more simple
that an external reference.
 

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #109 on: December 25, 2021, 12:49:45 am »
Merry XMas to everyone.
 
The following users thanked this post: MegaVolt

Offline Roehrenonkel

  • Regular Contributor
  • *
  • Posts: 167
  • Country: de
Re: DIY-SMU Project
« Reply #110 on: December 26, 2021, 01:59:14 pm »
Hello Dave and all,

thank you for your great work.
It would be an great addition for my tube-/valve-curve-tracer and messurement-system.
But 2U hight by 9,5" would be a waste of space in my 19"-Racks. ;->
1360013-0
The 2 spares are for propper ventilation, but maybe i can get ridd of the HP 66332A.
A two-channel unit would be okay with extended range (+-350V max. 30mA).
Do you have plans in that direction??
 
Here some thoughts on the output-connector:
Triax is too hard to come by and too expensive.
BNC (maybe the HV-version) can't run the guard-signal propperly.
Mixed-D-Sub (5W5 or 9W4) with HF-contacts and RG188 PTFE-cable will be my choise.
Inner conductor signal, outer conductor guard and the shell for shield.
Would have to assemble a "snake" out of 4* RG188, control-wires, shield-braid
and outer isolation.
Control-wires can be safety-switch, hardware-/manual lockout etc...
Maybe one could recycle some old high-quality VGA-monitor-cable
(at least 3 koaxes for RGB inside).
It's a compromise i know, but it's just one connector/cable to run to the DUT-fixture.

ADC-sampling: I see in "SPI.INO" that there is some hand-tuning the timing.
Didn't get the reason why, isn't the output-data-rate responsible for the stop-band frequencies?
Line-frequency should be selectable or even programmed via SCPI ("CAL:LFR 50" on my continent)
and the ADC-timing set acordingly for best noise-rejection.

Just some thoughts Dave, kudos for your work.

 

Offline ducreux92

  • Contributor
  • Posts: 17
  • Country: fr
Re: DIY-SMU Project
« Reply #111 on: December 30, 2021, 10:12:00 am »
Hi,

What is it L2 "1000ohms" : is  an inductance, value ?
« Last Edit: December 30, 2021, 03:21:08 pm by ducreux92 »
 

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #112 on: January 03, 2022, 09:59:35 am »
I don't see an L2 in my schematic MainBoardr Rev2b. However, there is an L1 or an FB2. L1 being a differential choke that mainly eliminates common mode noise present on the 12V input to/from the DC-DC converter.
FB2 is a ferrite bead. Those components exhibit a frequency dependant, strongly non-linear resistive behaviour. In your example "1000 ohm" would mean that @1MHz the -mainly resistive- impedance of the componenet is about 1k. Thi makes it ideal to make a low-pass RC filter to suppres -again- HF noise on a power line.
 

Offline Roehrenonkel

  • Regular Contributor
  • *
  • Posts: 167
  • Country: de
Re: DIY-SMU Project
« Reply #113 on: January 03, 2022, 04:28:39 pm »
Hi RikV,
 
....have a look again, Page 1/2 lower right corner Pin 20 of the ADC (Avdd).
L2 1000 ohm
In my guess it could mean 1uH (microhenry).
 

Offline 8334455

  • Newbie
  • Posts: 3
  • Country: us
Re: DIY-SMU Project
« Reply #114 on: January 04, 2022, 01:15:58 am »
Hi , any update about your hardware docs? and when ? expecting new awesome work!
 

Offline ducreux92

  • Contributor
  • Posts: 17
  • Country: fr
Re: DIY-SMU Project
« Reply #115 on: January 06, 2022, 01:47:41 pm »
hi,

1 uh = 6 ohms only (2*3.14*1E-6 * 1E6)

not 1000 Ohms at 1Mhz
 

Offline Roehrenonkel

  • Regular Contributor
  • *
  • Posts: 167
  • Country: de
Re: DIY-SMU Project
« Reply #116 on: January 07, 2022, 12:17:06 pm »
Hi,
 
maybe Dave (Ericson) wants to clear the "L2 1kOhm-mystery" up? ;-)

Reminds me of the time (1920s) when capacitor-values were in cm (1cm ~1.1pF)

Some alternatives for the Voltage-reference:
MAX6325 is stable with all c-loads,
LT1019-2,5 is stable with >2uF (+2 Ohms ESR like in Jaromirs J-SMU design).
or the Apex VRE3025 (0.6...1ppm Drift per °C), but don't know yet about the stability.

Got an answer from Apex now:
"The maximum capacitive load that VRE3025 can handle is not specified / known.
BUT:

As long as your input voltage to the VRE3025 is less than 12V,
the output is protected against shorts to ground indefinitely.
So, a 10uF decoupling capacitor as a load would not be an issue then.
If your input voltage is 12V or larger, but smaller than 40V,
the output can sustain a short to ground for 5s.
I guess this would easily suffice to fully charge a 10uF decoupling as a load."

« Last Edit: January 18, 2022, 09:22:21 am by Roehrenonkel »
 

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #117 on: January 15, 2022, 03:51:52 pm »
Guys, google "ferrite bead" or read this: https://www.analog.com/en/analog-dialogue/articles/ferrite-beads-demystified.html

Dave, could you pls release the latest revision of the pcb's, schematics, software, GUI files? Eventually as-is? There are so many modifications to the main board V2 that it is impossible to "bodge" them in.
 

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #118 on: January 19, 2022, 03:50:18 pm »
Dave,
If you are still redesigning the main board, why not consider to replace U19.1, R51,52,53,55 by a single AD629 or AD8276? It is a more precise, more stable and cheaper combination than the existing diff.amp.
 

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #119 on: January 25, 2022, 05:32:37 pm »
I finally got mine working. That is: I can adjust FV and the MV is corresponding within 1%. current clamping works.
So I thought: let's calibrate the V-ranges before going into the FI functionality.
I had tested SCPI before and that was working up to spec.
But: when I start a SCPI session now the comm with the ADC blocks: no readout an SPI message "ADC not responding" or so and no more USB response. Everything else works: Force setting, clamping...
I remember Dave saying he had such a problem when he was trying to add the push-buttons. Timing errors on his "bit-banging" SPI.
Anyone else building this device wanting to share experiences? Does Teensy3.2 have SPI libraries available for its "native" SPI port?
Any suggestions, help?
Correction: USB works since I get the error message.
It is even "worse" than that: the ADC ALWAYS crashes, even when the instrument sits there untouched. Sometimes after 2 hours, sometimes after 5 minutes.
I will try to implement SPI through SPI lib.
« Last Edit: February 03, 2022, 11:55:11 pm by RikV »
 
The following users thanked this post: Roehrenonkel

Offline Roehrenonkel

  • Regular Contributor
  • *
  • Posts: 167
  • Country: de
Re: DIY-SMU Project
« Reply #120 on: February 17, 2022, 11:08:01 am »
Hi RikV,

....why not consider to replace U19.1, R51,52,53,55 by a single AD629 or AD8276?
It is a more precise, more stable and cheaper combination than the existing diff.amp.

Yes, i had the same idea for the current-sense-amp (U21.1, U21.2, R47,50,56,59).
In that position the INA101 is the only choise.
Btw: Most INAmps have horrible input-offset-specs. for todays standards.

Another solution could be the LT5400 matched resistor-arrays.
Too bad they cost more than a complete InAmp.
« Last Edit: February 24, 2022, 11:51:51 am by Roehrenonkel »
 

Offline Roehrenonkel

  • Regular Contributor
  • *
  • Posts: 167
  • Country: de
Re: DIY-SMU Project
« Reply #121 on: February 18, 2022, 10:33:48 am »
Hi,
 
here is the missing PDF of the CPU-board.

Programming the Teensy from my Raspberry was no problem.

But the upload to the Nextion-TFT (NX4832T035_011) via SD-card doesn't work for me.
Tried different SD-cards with different formating (FAT/FAT32).
Maybe my TFT-file is corrupt? I have no propper Windows-PC, so i can't use the serial-upload.
If someone wants to check the file it's attached.

Update 2022-2-23: Raspberry FAT32 seems to be different to Windows FAT32.
Anyways the upload to the TFT works now. :-)
« Last Edit: February 23, 2022, 08:08:28 pm by Roehrenonkel »
 

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #122 on: February 20, 2022, 11:17:13 pm »
Looks like I finally got it working in a stable way, everything running on "real" SPI with the complete keypad decoding.
Tomorrow I will let it run for a couple of hours and then I'll callibrate it.
After cleanup I will post the post the soft to Dave, hoping it will inspire him to continue on the project.
Since I still have to disassemble the CPU board I think I will add an SPI_SS3 J2-pin12 to Teensy I/O3(pin5). On the main board there is a free input on U25. 2pieces of wire-wrap will bring a SS to uninstalled EEPROM U6. Nice to  save cal. factors.
Unless Dave had the intention of using this spare pin for something else??
« Last Edit: February 20, 2022, 11:24:41 pm by RikV »
 
The following users thanked this post: Roehrenonkel

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #123 on: February 23, 2022, 08:33:27 pm »
Does any body know if Dave is OK? It is so very quiet from his side. I hope he and his family are fine.
 

Offline RikV

  • Regular Contributor
  • *
  • Posts: 130
  • Country: be
Re: DIY-SMU Project
« Reply #124 on: February 23, 2022, 10:42:52 pm »
I tried to upload my files to Github but it did not work, so I post them here. Maybe some good soul uot there could move them to the right place?
I included the SPI version of the *.ino, used libraries and my Nextion file. Hope this will be of some help.

I have to split the upload.
 
The following users thanked this post: Roehrenonkel


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf