• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Clamping output to gnd at very low threshold?

Thread starter #1
I have a digital potentiometer setup as voltage divider that doesn't quite go down to 0V and can only go as low as 30mV. Is there any logical way to clamp this to GND when it is at this level, for example, using a combination of analog switch / comparator or likewise? Would like it to be adjustable since these DACs have quite a bit of tolerance.

thanks
 
Last edited:

dknguyen

Well-Known Member
Most Helpful Member
#2
You mean like a giant sub-mohm MOSFET wired to pull down the output of the digital pot whose gate is driven by a comparator whose output fires whenever the input falls below ~30mV? If so, I kind of wonder how you would ever get out of that state since you have a feed forward loop where the MOSFET drives the output to zero and stays at zero if the output is driven to zero.
 

JimB

Super Moderator
Most Helpful Member
#3
A rather broad question with not much supporting information.

Which digital potentiometer? A link to its datasheet would be useful.

How are you using the digital pot? A schematic would be useful.

JimB
 
Thread starter #4
You mean like a giant sub-mohm MOSFET wired to pull down the output of the digital pot whose gate is driven by a comparator whose output fires whenever the input falls below ~30mV? If so, I kind of wonder how you would ever get out of that state since you have a feed forward loop where the MOSFET drives the output to zero and stays at zero if the output is driven to zero.
I am not sure how this can be done or if it is even possible, but you never know.
 

dknguyen

Well-Known Member
Most Helpful Member
#5
I am not sure how this can be done or if it is even possible, but you never know.
Well, it's a digital pot so your processor should know when it's trying to get 0V out of the pot. In which case, drive the MOSFET gate directly with the processor and use software to coordinate the MOSFET and the pot.
 
Thread starter #6
A rather broad question with not much supporting information.

Which digital potentiometer? A link to its datasheet would be useful.

How are you using the digital pot? A schematic would be useful.

JimB
I am using AD5292 with a 5V reference connected to it A terminal while B is connected to GND. W is connected as reference voltage to a 10BIT DAC.

Hope that helps
 
Thread starter #7
Well, it's a digital pot so your processor should know when it's trying to get 0V out of the pot. In which case, drive the MOSFET gate directly with the processor and use software to coordinate the MOSFET and the pot.
Actually, thats a good idea. Now you got me thinking that I can use one of the GPIO pins on my SPI interface to drive a switch.

Thanks a bunch
 

dknguyen

Well-Known Member
Most Helpful Member
#8
Actually, thats a good idea. Now you got me thinking that I can use one of the GPIO pins on my SPI interface to drive a switch.

Thanks a bunch
What MOSFET are you planning on using? MOSFETs take a finite amount of time to turn on because the gate capacitance needs to be charged. Especially very low RDson MOSFETs (high gate capacitance/charge) being driven by the weak, low-current GPIO of a regular IC. Don't forget to use a logic level MOSFET.

Of course, if you needed speed then you could drive the gate with a a dedicated low-side gate driver IC triggered from the GPIO. Which also releases you from the bounds of a logic level MOSFET since gate driver ICs have a separate supply voltage input used to drive the gate. But you probably don't need speed. Just putting it out there in case you did.
 
Thread starter #9
What MOSFET are you planning on using? MOSFETs take a finite amount of time to turn on because the gate capacitance needs to be charged. Especially very low RDson MOSFETs (high gate capacitance/charge) being driven by the weak, low-current GPIO of a regular IC. Don't forget to use a logic level MOSFET.

Of course, if you needed speed then you could drive the gate with a a dedicated gate driver IC triggered from the GPIO. Which also releases you from the bounds of a logic level MOSFET. But you probably don't need speed. Just putting it out there in case you did.
I am guessing like a simple analog switch like ADG1219 should work fine for my purpose.
 

Pommie

Well-Known Member
Most Helpful Member
#11
Something doesn't add up. Your maximum error should be 10mV (5V/1024 * 2) - the 2 is the max error at position zero. Your normal error should be zero!!

A schematic would help.

Mike.
 

JimB

Super Moderator
Most Helpful Member
#12
I just had a look at the datasheet to see if I could find the minimum resistance of the potentiometer, but I could not see it.

Rather than trying to short circuit the output at zero output, why not insert an offset adjustment, maybe using an op-amp, to take the residual output to zero.

JimB
 

ronsimpson

Well-Known Member
Most Helpful Member
#16
5V reference connected to it A terminal while B is connected to GND. W is connected as reference voltage to a 10BIT DAC.
Is the B terminal really connected to the ground reference of the 10 bit DAC?
If you short out the W terminal to ground of the "pot" what is the reading on the DAC?
----edited----
It is easy to have 30mV drop in ground in a bread board.
 
Last edited:
Thread starter #17
Is the B terminal really connected to the ground reference of the 10 bit DAC?
If you short out the W terminal to ground of the "pot" what is the reading on the DAC?
----edited----
It is easy to have 30mV drop in ground in a bread board.
Just tested this and the DAC output is Zero when I short it. When I remove the short it goes back to 30mV. This setup is on a PCB. B terminal is connected to GND layer.
 

Latest threads

EE World Online Articles

Loading

 
Top