Are these load cells half-bridge or quarter-bridge ?

Hello,

I have opened a cheap digital kitchen scale and want to connect it to an Arduino. It contains four load cells.

However I don't know whether they are of half- or quarter-bridge type. I guess they are of quarter-bridge type because of the way they are connected to the PCB (S+, E-, E+, S-).

Could someone confirm this ?

Please have a look at the pictures here:

Thank you in advance.

3-wires: Quarter-bridge.
4-wires Half-bridge.

You have Half-bridge sensor. E+/- are Excite - S+/- is Sense +/-

// Per.

I don't get it. The cells do have three wires. So how can they be half-bridge ?

amundsen:
I don't get it. The cells do have three wires. So how can they be half-bridge ?

The cells are "paired" 2&2.

// Per.

Surely a 2-wire sensor is quarter-bridge (1 resistive element), a 3-wire sensor is half-bridge (2 resistive elements), and a 4- or 6-wire sensor is a full bridge (4 resistive elements)?

The load cells in your photo appear to have 3 wires, so they are half-bridge types. Ther are probably paired to give two full bridges.

Interesting. Any idea of the way the signals from the two pairs are combined ? Averaging ?

amundsen:
Interesting. Any idea of the way the signals from the two pairs are combined ? Averaging ?

Summation. After all, the scale needs to display the total load on all 4 sensors.

Any schematics showing a configuration with such a dual full-bridge would be useful. I can't find any on the Web.

Also, despite the 6V power source, there's only 2.05 V between E- and E+. Therefore, I don't understand where the need for 6V (or for 5V more probably) comes from. The backlit LCD ? The microcontroller ?

amundsen:
Any schematics showing a configuration with such a dual full-bridge would be useful. I can't find any on the Web.

Also, despite the 6V power source, there's only 2.05 V between E- and E+. Therefore, I don't understand where the need for 6V (or for 5V more probably) comes from. The backlit LCD ? The microcontroller ?

The actual voltage applied to the E+/- is not so important as it is to be a stable 'reference voltage' as it sets measurement range (or sensitivity if you like). However it's important that you know the exact excitation voltage as it is needed to scale the voltage reading to a real world units measurement.

Actually I don't need to convert the output as a real world unit measurement. I only want to use the pressure as a control for musical purposes.

I understand that the actual excitation voltage is not very important but still I wonder why it's so low with a 6V power as the output voltage of a load cell is so small yet. Also, as I want to connect the bridge to a 5V reference I want to be sure not to burn the load cells.

After some voltage and resistance measurements I have an idea about the way the sensors are connected together.

However I am not sure about it because of the resistance measurements. The resistance values shown in the picture are made when the circuit is not powered. When powered, I have differents figures. Instead of 880?, I measure -8k? but if I swap the probes, the measure is out of range (> 20k). Also, if I try to measure a resistance variation according to pressure, I can see the resistance between points 7 and 9 or points 4 and 6 evolving around 1.5k, but I cannot measure any similar figures on the two other sensors: I got a mugh higher value but with a negative polarity again (-11.25k?) between points 1 and 3 (+ probe on 3 and -probe on 1) or 10 and 12 and when swapping the probes the value is out of range.

Do you think it's normal and possible to have such figures with this schematics ?

You can't use a multimeter to measure resistance when the circuit is powered (well you could, but it would require a special kind of multimeter). The resistance change when you load the sensors will be very small, so I am not surprised that you can't measure it. The sensors are wired in a bridge configuration, so that a small change in the resistances will cause a relatively large change in the voltage between S+ and S- (relative to its no-load value, that is).

dc42:
You can't use a multimeter to measure resistance when the circuit is powered

Aha. :fearful: :blush: XD

Thanks dc42 !