Resistors are installed, connected and trimmed for close nominal value.
I've decided to solder 40kΩ elements to thick copper sheet to slow down them thermally. Will see how that will work :-).
All connections are done with multistranded unshielded copper wire, crimped to copper lugs. Green post on the side is for grounding purposes and twisted white/black cable is PT1000 sensor.
10 kΩ output formed out of four VHP101T 40kΩ in parallel. Each of these is about 91 USD on Digikey so not the cheapest way :-)
Pins from VHP-4 1 ohm element stick up via heatsink openings. It will be easy to connect wires this way and reduce amount of moving around during assembly.
Everything wired up:
First check of 1 Ω output with 3458A meter. A little bit high, so I can trim it down with parallel resistor across sense terminals.
First check of 10000 Ω output with 3458A meter. A little bit high, so need to trim that down.
Trimmings for nominal, 88 MΩ using PTF56 resistors for 10kΩ and about 500 Ω BMF for 1 Ω output.
Now both outputs are closer to desired outputs.
Next test is with Measurements International 6010B DCC bridge for 10 kΩ output stability.
Live data page available here.
Trim added for 10000 Ω output, and RTD HEL-705 1000 Ω temperature sensor wired up and connected to 3459X DMM to monitor resistance sensor. URL with data
I tried to get 1 Ohm to better TCR but it was already rather trickly and behind the schedule for friend use, so I've left it as is.
Thanks to integrated temperature sensor it's not hard to compensate for different box temperature to obtain accurate resistance value with good uncertainty.
I'm interested in what diecast box was used with the nice black powder-coating. Black colour would allow for a professional looking front panel to be made out of a PCB.
This is what I did for my 10k standard, but it clashes with the grey box that I have. I'll post pics when I'm not on in-flight internet. I did steal the temperature sensor connector and pinout from the your box (though I used a PT100, as it was easier to get from aliexpress already with a 4 wire lead).
In this topic, I'll post information about the build of the DIY resistance standard boxes to help out a friend.
Will be building box with two outputs.
Output 1
Nominal resistance: 10000 Ω ±4ppm
Calibration uncertainty: ±0.3 ppm
Desired TCR: <0.5 µΩ/Ω
Terminals: four-wire TBP-1
Temperature sensor: PRT 1000
Target cost: 300 CAD
Output 2
Nominal resistance: 1 Ω ±4ppm
Calibration uncertainty: ±0.5 ppm
Desired TCR: <2 µΩ/Ω
Terminals: four-wire TBP-1
Temperature sensor: PRT 1000
Target cost: 300 CAD
Target resistors:
VHP for 10k , VHP-3 for 1 ohm
Got some enclosures for a build.
Drilled holes and installed posts. Prepared 1 ohm unit to install.
Resistors are installed, connected and trimmed for close nominal value.
I've decided to solder 40kΩ elements to thick copper sheet to slow down them thermally. Will see how that will work :-).
All connections are done with multistranded unshielded copper wire, crimped to copper lugs. Green post on the side is for grounding purposes and twisted white/black cable is PT1000 sensor.
10 kΩ output formed out of four VHP101T 40kΩ in parallel. Each of these is about 91 USD on Digikey so not the cheapest way :-)
Pins from VHP-4 1 ohm element stick up via heatsink openings. It will be easy to connect wires this way and reduce amount of moving around during assembly.
Everything wired up:
First check of 1 Ω output with 3458A meter. A little bit high, so I can trim it down with parallel resistor across sense terminals.
First check of 10000 Ω output with 3458A meter. A little bit high, so need to trim that down.
Trimmings for nominal, 88 MΩ using PTF56 resistors for 10kΩ and about 500 Ω BMF for 1 Ω output.
Now both outputs are closer to desired outputs.
Next test is with Measurements International 6010B DCC bridge for 10 kΩ output stability.
Live data page available here.
Seems like 10000 Ω output is somewhat a thermometer :-)
Time to rectify that with couple trims to the network and try again?
Trim added for 10000 Ω output, and RTD HEL-705 1000 Ω temperature sensor wired up and connected to 3459X DMM to monitor resistance sensor.
URL with data
Final box outlook and calibration results.
I tried to get 1 Ohm to better TCR but it was already rather trickly and behind the schedule for friend use, so I've left it as is.
Thanks to integrated temperature sensor it's not hard to compensate for different box temperature to obtain accurate resistance value with good uncertainty.
I'm interested in what diecast box was used with the nice black powder-coating. Black colour would allow for a professional looking front panel to be made out of a PCB.
This is what I did for my 10k standard, but it clashes with the grey box that I have. I'll post pics when I'm not on in-flight internet. I did steal the temperature sensor connector and pinout from the your box (though I used a PT100, as it was easier to get from aliexpress already with a 4 wire lead).