Quartet Of SMD Resistors Used To Sense Z-Axis Height | Hackaday

This is the clever trick for your next 3D printer construction or transformation:

. We bet it may have many other applications.

Traditional load cells, at least those you can pick up cheaply from usual sources or harvested from old kitchen or bathroom scales, are usually too large to be used on the extruder of a 3D printer. [IvDm] Want to make a touch sensor for it

, So he built his own load cell to complete this task. It is composed of four 1000 ohm SMD resistors with 25 larger device sizes. He mounted them on an X-shaped PCB and wired them in a classic Wheatstone bridge configuration with two resistors on one side of the board and two resistors on the other side.

The extruder is installed in a hole in the center of the board and floats on the board. Through the HX711 load cell driver chip, when the extruder hits the bottom of the bed, the bridge can sense the slight bending of the board, and ATtiny85 pulls the limit switch input to the ground. [IvDm] even performed some repeatability tests on this sensor, and the results were unexpectedly consistent. About the first minute of the following video shows how it runs on Hypercube.

We found that the use of SMD resistors as strain gauges here is very clever, but it has many relationships with off-the-shelf load cells:

,

, Even

.

Thanks [jschoch] for the hint.

How does it work? Is it related to the carbon in it?

This is really interesting.

Damn, this is so smart. I understand how it works, but this is one of the things that I may not consider impractical. I am surprised that it actually works.

If you understand how it works, why are you surprised at how it works?

This is like the experiment of ether wind, or the complex, broken and hardened stone blade made by early humans. You know it is possible, but it can work well for a small amount of money.

Because although it can work, the desired effect is still one of the criteria for resistors specifically designed. They should remain stable under temperature, aging and mechanical stress. Changing its value through mechanical input is usually called a microphone. I bet that the cheaper the resistor, the better the effect.

Durfus mounted (intentional) resistors are usually film based, so if you engrave a zigzag conductive pattern on the film (thin or thick), compression or stretching will change the length of the pattern and change the resistance

But only one point... although it can be measured

Osgeld

Zigzag is not really relevant: it is a method of impedance matching and has the side effect of introducing anisotropy. Fortunately, most SMT resistors are not made this way (except for some ultra-high resistors).

Almost all SMD resistors (especially in the 1 kohm range) are metal films. But no matter what, it works by bending the resistor to stretch the metal film, thereby increasing its resistance. The number is small, but it is enough to measure.

This is a clever design! I want to know whether the designer has calibrated the output volts/mm deflection to understand the sensitivity/repeatability or possible stress on the solder joints?

One possible way to get higher sensitivity from this design is to place a small groove on the PCB directly below each resistor. The slot will slightly weaken the area of ​​the PCB, thus concentrating the bend here instead of spreading it over the entire arm. At least it's worth a try.

*route*

Drill a hole

Mouse bites?

"Rout" is correct. You use a router to eliminate a slot. "Route" is a path.

No.

Rout-verb-hollow out or cut out grooves or patterns on the surface

Moreover, the definition of "route" does not include this definition.

Or you can use strain gauge

They even have similar dimensions to resistors.

I guess they should start making SMD strain gauges :-)

Yes, but these resistors cost less than 10 cents. Even cheap strain gauges are only about $1.

It would be great if something like this can be proven to be sufficiently repeatable and robust for a given application.

They may be cheap, but due to the stiffness of the ceramic substrate, they are not sensitive.

Agree with the clever "why didn't I think of that" thing.

A quick look at the linked files did not find out whether they use thin film resistors or thick film resistors, but I guess the thin film would be better here (lower noise, higher accuracy, lower temperature coefficient, metal and oxide resistor elements ).

But now I think about it, why not just use bent copper traces as elements on the PCB? Using the 6 million rule, you can easily get a 2-inch trace in the space of 2512 components. The resistance is still low, but the noise is low, and the mechanism is not bothered by solder resist and very hard ceramics for resistors.

If a proper (low-Z) amplifier is used, I bet the sensitivity and noise performance of the PCB trace will be better than resistors.

At least for Delta printers.

The designer even puts the driver circuit on the effector.

I have not seen the PCB trace variants of Cartesian printers, which may be due to the variability of the installation style.

Does anyone order PCB from JLCPCB for this? I will spend some money-the minimum order quantity is 10pcs, I will choose 2 :)

What's really surprising is that the parts they make are actually designed for this service, called strain gauges. I suspect that they are not as cheap as SMD resistors, but in most cases, I like to use parts that are designed for me to use them.

The first problem that ensues is mechanical creep in the solder joints, which is under pressure in this design. When using a strain gauge, the strain gauge itself is flexible and does not bear mechanical forces-it only deforms with the base material, and is designed to do so so as not to interfere with the mechanical device.

Therefore, the most likely failures after multiple load cycles are: a) The solder joints begin to conform to the load, and the zero point of the bridge gradually deviates from alignment; b) The solder joints produce whiskers due to RoHS soldering and mechanical strain. , Eventually leading to a short circuit; c) The solder joint is broken.

As long as you don't want to sell the device, you can easily solve the points b and c: avoid the use of of-pin (RoHS) solder. But I also prefer to use suitable components at work, in this case strain gauges.

You can of course use the zigzag traces on the PCB itself to do this.

(But don't you want to measure the height of the extruder without pushing it into the bed?

I think capacitive or inductive induction is better. )

I often wonder why people don’t just make the bed conductive, but use a pogo pin probe. Effectively replace the sensor with a switch and perform the measurement directly.

This is an example of a PCB-based load cell.

Thank you. It’s great that my main concern is the space required for this. Cca. The top and bottom layers are 12 square centimeters. I don’t know if this is useful without an op amp.

It's not like how much money has been spent on Ali for load cells and UV resin these days...

But yes, I think this can also work :-)

You can also just make a PCB with tiny swing lines and measure it.

Please be kind and respectful to help make the comment section great. (

)

The site uses Akismet to reduce spam.

By using our website and services, you expressly agree to our placement of performance, functionality and advertising cookies.