Voltage divider protected by Zener diode

[pgib8]

I'm looking for a voltage divider that also has its output clipped at about 3V.
I need to do voltage measurement and I'm going to use a voltage divider. I want to "magnify" the lower region.
The input is 0 to 20V but I'm really interested in 0 to 4.2V (Li-Ion battery).
So I thought I use a divider for the region of interest and clip the rest.
R1 is 6,000 Ohm, R2 is 10,000 Ohm.
The Zener I'm considering is DDZ3V0BSF-7 (Diodes Incorporated).
The Zener voltage is 3.01V to 3.22V, let's assume it is 3.1V.

Unfortunately I don't think this will work as it is.
At 4.2V at the input, the output is 2.6V. So far so good, but the current will be only 0.4mA. The Zener is already letting this amount of current pass through at 2.6V, and this will mess up the measurement.



I assume to fix the problem I will need more current coming through R1 and in addition re-scale the divider and say that anything above 2.5V will already have significant leakage through the Zener diode.
I also can't make R1 too small because at 20V I have to keep in mind how much power the Zener can clamp before burning up (500mW).

Any clever ideas or thoughts on how to do this very efficiently? The current consumption doesn't matter because it is only powered periodically.

As for my own idea, I'm considering two voltage dividers in series and the zener coming between them.

 

[Ratch]

I'm looking for a voltage divider that also has its output clipped at about 3V.
I need to do voltage measurement and I'm going to use a voltage divider. I want to "magnify" the lower region.
The input is 0 to 20V but I'm really interested in 0 to 4.2V (Li-Ion battery).
So I thought I use a divider for the region of interest and clip the rest.
R1 is 6,000 Ohm, R2 is 10,000 Ohm.
The Zener I'm considering is DDZ3V0BSF-7 (Diodes Incorporated).
The Zener voltage is 3.01V to 3.22V, let's assume it is 3.1V.
View attachment 21437
Unfortunately I don't think this will work as it is.
At 4.2V at the input, the output is 2.6V. So far so good, but the current will be only 0.4mA. The Zener is already letting this amount of current pass through at 2.6V, and this will mess up the measurement.

View attachment 21438

I assume to fix the problem I will need more current coming through R1 and in addition re-scale the divider and say that anything above 2.5V will already have significant leakage through the Zener diode.
I also can't make R1 too small because at 20V I have to keep in mind how much power the Zener can clamp before burning up (500mW).

Any clever ideas or thoughts on how to do this very efficiently? The current consumption doesn't matter because it is only powered periodically.

As for my own idea, I'm considering two voltage dividers in series and the zener coming between them.

I don't know what you are trying to do, but I assume you want your output to be a constant voltage. It seems to me that R2 is just another resistive load in parallel with the load. I don't think you want that, so throw away R2. Calculated the max current the diode can handle from its wattage rating. Then calculate R1 to limit the current for a no load situation at the highest voltage, which stresses the diode the most. That is the best you can do for a unstabilized input voltage, unless you can change R1 every time the input voltage changes.

Ratch

 

[pgib8]

I don't know what you are trying to do

I'm building a voltage divider to scale the input voltage down to something that the 10-bit ADC can measure, to do voltage measurement.
I want to have the maximum resolution in the area of interest (3 to 4.2V) and I don't care about anything above that. The max input voltage is 20V.
If I just scaled it down so that 20V equals 3V at the output, I would have a low and marginal resolution of 0.02V per ADC-step.
I am utilizing the Zener diode to clip the input voltage, which allows me to re-scale the voltage divider so that I can get a higher resolution for my battery monitoring (3 to 4.2V).
I have adjusted my circuit and I'm now happy with it. The resolution is now 0.01V.
I have selected a different Zener diode: (Digi-Key # BZT52C10-TPMSCT-ND)


Note1: The leakage current is now completely negligible, which was the issue in the first design.


I should add that this is not continuously energized, only during voltage measurements.

 

[CDRIVE]

If you'd like to bring the clamping voltage a bit closer to your desired 4.2V you can add a 1N4002 Diode in series with the 3V Zener. The clamping voltage will then be ~ 3.7V. Two series Diodes will bring your clamp voltage up to ~4.4V.

Chris

 

[pgib8]

@CDRIVE, that's a good tip. Using the forward voltage always seems like cheating but if it works, it works. In my case I've gone to a 10V Zener and the voltage divider comes after it.