My personal Scale project has a huge drift in its readings for weight

[drmina2023]

for the last few months, I was developing a device measure personal height and weight, and I almost finished it as I designed a chassis to host all the electronic components.
here is the final chassis

the Idea is simple I bought a commercial scale with 200 kg load cell but without any screen.
I connected the 4 wires from the load cell to HX711 ADC that is connected to an Arduino nano and an Lcd here is my electronic casing.

and here is the scale

all the electronic components including the HX711 are placed in a plastic box that is situated at about 140 CM from the load cell and I used an insulated breaded wire with a metal shield to connect the load cell to the HX711 to reduce electronic noise as you can see here
breaded wire - Search Images (bing.com)
here is the wiring diagram

the problem I'm facing now is that when I measure my weight it gives a correct measurement at first, but the reading starts to decline within minutes to be decreased by about 7KG in 30 minutes, so it starts at 96.6 to reach 89 kg in 30
minutes.
Any Ideas how to stabilize the readings ?

[gilshultz]

Can you measure the actual voltage of the load cells? If so, do the values change? You can also replace the load cell with a dummy load to see if it drifts. This will help you identify if the source of the drift is in the load cell or elsewhere.

[drmina2023]

I'll try to measure the voltage , but for the dummy one I think it's not available,
In addition the scale is a commercial one that works greatly with it's dedicated Screen.

[johnerrington]

@drmina2023

I connected the 4 wires from the load cell to HX711 ADC that is connected to an Arduino nano

it may seem trivial but can you provide a simple schematic including the load cell(s), the SparkFun Load Cell Amplifier - HX711 - SEN-13879 - SparkFun Electronics

and the nano and its power supply

How to make a schematic you can post.

When you submit a question on the forum you will usually need to add a schematic (circuit diagram). The tutorial that follows will explain some ways to create a useful image; however lets start by explaining how to add the image to your post. Its easy. Having created the image all you need to do is : Save the image somewhere convenient - maybe your desktop Start your post, and place the cursor (insertion point) where you want the image to be shown; [upload image to forum] 3: click th…

[DaveX]

Sounds mechanical to me.

If you step off after the 7-minutes drift, does the unloaded reading have an offset that also drifts back to zero?

@drmina2023

a commercial scale with 200 kg load cell but without any screen.

@drmina2023

In addition the scale is a commercial one that works greatly with it's dedicated Screen.

Do you have a datasheet on the unnamed commercial scale?

My commodity bathroom scale has a definite measuring cycle: finding a stabilized reading, timed holding, and then shutting off to encourage you to step off before a new cycle. Maybe they plan the timing cycle to limit the effects of drift?

[sonofcy]

I find that is normal behaviour for a load cell. It's not critical for me so I have not investigated, but my hunch is some of the more esoteric functions of the HX711 library may address it.

[EmilyJane]

Similar question to others: Is this a "zero" shift in that when you step off the scale the reading doesn't return to zero or is a "gain" shift in which the apparent gauge sensitivity seems to have changed?

[DaveX]

Sparkfun shows specs on their 200kg load cell as +/-0.024%FS/30 minutes.

7kg/200kg in 30 minutes is 3.5%, which is relatively huge.

[drmina2023]

@EmilyJane
it does return to zero after step off the scale , could you please check my sketch here

[EmilyJane]

@drmina2023

does return to zero after step off the scale

I would guess then that either the gauge factor of the load cell is changing over time or the gain of the HX711 amplifier is drifting. Since it's a commercial unit, I would suspect the HX711 might be the culprit. Post #2 is good advice.

[lastchancename]

Power supply ? / battery ? & schematic?

[drmina2023]

@lastchancename
@johnerrington
I'm using the nano USB to power the whole project
here is a simple schematic

note that the actual thing has a 140 cm breaded wire to connect the HX711 to the load cell
and here is the full schematic

[johnerrington]

@drmina2023

when I measure my weight it gives a correct measurement at first, but the reading starts to decline within minutes to be decreased by about 7KG in 30 minutes, so it starts at 96.6 to reach 89 kg in 30
minutes.

You remain on the scale for 30 minutes?

@drmina2023

I'm using the nano USB to power the whole project

Have you measured the resistance of the load cell(s)? Perhaps the current draw is too much for the supply?
Do you have one of these to check the current drawn from the USB?
https://www.aliexpress.com/item/1005006550950357.html?src=google

[drmina2023]

no, what I mean is that the I made a measurement every few minutes and every time the reading is getting less each time.

[lastchancename]

You’re hanging quite a lot off the Nano 5V rail… I’d be looking at a second power source for all the peripheral supplies.

[drmina2023]

Could you please, suggest me how to do that

[lastchancename]

I’d be looking at a second power source for all the peripheral supplies.

[johnerrington]

Did you miss this?

@johnerrington

Have you measured the resistance of the load cell(s)? Perhaps the current draw is too much for the supply?
Do you have one of these to check the current drawn from the USB?
https://www.aliexpress.com/item/1005006550950357.html?src=google

[drmina2023]

I'll do the measurements soon and give you a feedback as soon as possible .

[hammy]

Also worth checking the base resistance of the scale in case it is not suitable for the HX711 -
Very unlikely , but maybe the load cell is warming up?
Beware wiring layouts too - 0v and power leads should all be star connected to a single point .
Also check the load cell is isolated from the platform.

Nice project !!