EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: Arsh ahmad on August 17, 2024, 07:16:10 pm
-
Hi all,
I saw the FLIR decade resistance Box, So I have built a similar using KiCad, which used SPDT switches and 1206 Resistors.
Designs can be accessed from below links:
[/https://github.com/tslistener-Arsh/Decade-Resistance-BOX (https://github.com/tslistener-Arsh/Decade-Resistance-BOX)]
!! update - SPST Version Also Added
It uses only 1s Range of resistance values since it is hard to get all the values in smaller quantities.
I was a KiCAD guy but I have been using Cadence Suite in my organization. I am amazed that KiCAD is head to head with cadence with KiCAD 8.
-
Cloning the git repo and opening it in KiCad shows the configuration of this thing. There are many projects on the 'web Just on gitlab you can find almost 30.000 on github alone.
https://devbisme.github.io/RepoRecon/?topic=kicad&filter=&sort=pushed:desc (https://devbisme.github.io/RepoRecon/?topic=kicad&filter=&sort=pushed:desc)
Without some more info about your project and more accessible data such as a schematic in .png. .svg or .pdf format or a picture of the actual finished product, very few people will be bothered to go though the effort to see whether your project would be worth something to them. Without a better description on github it is already difficult to even find your project in between all the others.
About the project itself. I would like a higher resolution, for example for simulating the effect of temperature changes on an Pt100 you need quite small resistance changes. Adding more copper to act as a bit of a heatsink also helps with cooling the resistors.
-
What is the logic behind using SPDT switches to switch between a resistance and a short, instead of SPST switches to short out each resistor group?
-
Thanks for the feedback, sure I will release an update to the initial commit soon.
-
What is the logic behind using SPDT switches to switch between a resistance and a short, instead of SPST switches to short out each resistor group?
Yes, that SPST are the first choices to go, I had quite a few SPDTs Laying around so I though to use them for few resistance simulators , instead of getting new ones.
Also That parallel contact resistance of the SPST switch itself can be an issue in smaller values.
-
Also That parallel contact resistance of the SPST switch itself can be an issue in smaller values.
That problem is worse with your SPDT circuit.
No matter how many resistors are selected, the circuit path is always through 24 switch contacts.
By using SPST config, you reduce that to 23 switch contacts max.(One resistor selected) But for each additional resistor you select, you eliminate one switch contact resistance. Not a big improvement, I'll admit. :D
If you're really worried about the low resistances being accurate, you could add another switch between SW4-1 and J1 for those situations when you want R < 10 ohms.
Or build a separate low resistance box with fewer selections, or use a different switching method entirely.
-
One additional problem you should think about is that normal SPDTs is that they are break before make. That means the switch opens the connection before close the other connection is done.
In your circuit, if you operate a switch, the resistance goes to infinite for short time. Usually that is unwanted. That doesn't happen if you only enable a bypass at the resistor stages. Just use your SPDT as SPST switches.
-
Also That parallel contact resistance of the SPST switch itself can be an issue in smaller values.
That problem is worse with your SPDT circuit.
No matter how many resistors are selected, the circuit path is always through 24 switch contacts.
By using SPST config, you reduce that to 23 switch contacts max.(One resistor selected) But for each additional resistor you select, you eliminate one switch contact resistance. Not a big improvement, I'll admit. :D
If you're really worried about the low resistances being accurate, you could add another switch between SW4-1 and J1 for those situations when you want R < 10 ohms.
Or build a separate low resistance box with fewer selections, or use a different switching method entirely.
I was thinking of making custom switches using 3D Printed parts and PCB, I saw Marco Reps on youtube making a relay of his own using the same. Since all the 24 switches are in series the total resistance of the switch in series will be same and will not change based on value selected. This was also one of the reasons going with the switches.
-
One additional problem you should think about is that normal SPDTs is that they are break before make. That means the switch opens the connection before close the other connection is done.
In your circuit, if you operate a switch, the resistance goes to infinite for short time. Usually that is unwanted. That doesn't happen if you only enable a bypass at the resistor stages. Just use your SPDT as SPST switches.
Yes, this will be a problem when stepping the Values in a Hot circuit, But in the old school rotary switched based devices, they too had to break the connection before changing the values. Its is better to go back to design once again and see how this can be made better than just any other decade resistance box.
-
Why not just bridge the resistor with the switch. There's no need remove the resistor if you bypass it. Something like I crudely drawn in the attachment.
-
Yes, I am also working on a smaller compact design which uses DIP SPST switches and is smaller in size, that uses this method, I still have to precisely calibrate for the effect caused by resistance of the switch.
-
I have made a updated version that uses 4 Channel SPST DIP switches, I have also given option for Pin Headers and Jumpers instead of SPST switches. This one is also smaller in size , also I have added a port for ground plane.