Resistive equivalent of an R2R ladder?
Simon Merrett wrote 08/24/2018 at 17:52 • 0 pointsI know, an R2R ladder is made of resistors. But its purpose is to allow a stepped range of voltages (not resistances) to be produced at one output from only two input voltages, inserted at multiple points.
I'm trying to produce a variable resistor that can achieve 6 bit linear resolution (as opposed to log or some other curve - I want to slide up and down a ramp of resistance). My available method of doing this is to tie any point in a resistor array or network to a common rail (using a microcontroller to switch the GPIOs between inputs and outputs and then go to sleep). The terminals of this device, across which resistance would be measured don't have to be connected to that bus. I'd consider using other components but I'm assuming that resistors are what's needed.
I have looked online, looked for electronic decade resistors, looked in The Art of Electronics and it doesn't seem to have been done. But intuitively it seems like it should be possible.
Can anyone please give me some ideas about whether this is possible and where to look next?
Discussions
Become a Hackaday.io Member
Create an account to leave a comment. Already have an account? Log In.
SPST / SPDT, e.g. http://www.vishay.com/docs/62749/dg1411.pdf
I've encountered resistor selection with SPST ICs more than once in detector preamps, the advantage being that they're uncommitted within what the supply voltages permit.
And here's another fun thing to do with jFETs (although their range is too limited for your purposes)
https://www.vishay.com/docs/70598/70598.pdf
Are you sure? yes | no
MCP41010-I/SN , MCP42010-I/SN
Are you sure? yes | no
1uA static operation - awesome, thanks so much!
Are you sure? yes | no
digitally-controlled resistors (potentiometers) are a thing. You can set them by a variety of means, including I2C, SPI, rotary encoders, and parallel bits. Digikey has over 6,500 different ones listed (but some are just packaging variations, etc).
https://www.digikey.com/products/en/integrated-circuits-ics/data-acquisition-digital-potentiometers/717?k=digital%20potentiometer
They commonly do exactly what you mention, except using MOSFETs to switch fixed resistors in and out of the circuit. They're so cheap and convenient that you would probably never make one yourself, unless you just wanted the experience.
Are you sure? yes | no
Hi @Ted Yapo thanks for the suggestion. I've used digital pots before (and please forgive the feature creep here) I'm keen to specify the resistance range quite exactly (although there may be a work-around for this). The other preference I have is is for uA supply current when not communicating or switching wiper setting. The ones I have looked at are in the mA range so had ruled digital pots out. However, if low quiescent current ones exist I'm all ears!
Are you sure? yes | no
If you really need that level of "exoticness," how about a set of resistors switched by latching relays? As long as the smallest resistance step is much larger than the contact resistance, your precision should be limited by the resistor tolerance. This will consume zero quiescent power (except what's dissipated in the resistors). I built a variable RF attenuator like this many years ago. Some "high-end" audiophile volume controls also use such relay-switched resistor networks in stepped attenuators because normal pots sound objectionable to the true audiophile. In that case, you have to put up with the relay noise while changing volume levels, though.
Just make sure you get relays that can be switched dry (no current through the contacts). Most relays require you to switch a load to perform reliable.
Are you sure? yes | no
Funny enough a poti and a 9g servo would fit this description...
But I guess you don't want to spend the build volume and the huge amount of current to change the resistor.
Are you sure? yes | no