Welcome! Log In Create A New Profile


Why strain gauge?

Posted by leadinglights 
Why strain gauge?
May 18, 2023 08:17AM
I have noticed lately that there are a lot of printers using strain gauges for bed leveling, typically with HX711 or HX717 conditioner/ADC ICs.
My question is, Why? It is not the use of strain gauges themselves that is my concern: Strain gauges can be stunningly sensitive, fast responding, and mechanically stable for long periods - and they can be cheap as well. My problem is that almost all of the implementations I have seen used very high contact force, often more than 500 grams: Videos on YouTube frequently show detectable vertical movement of the bed when the nozzle contact is made.
I accept that there are added attractions in using a strain gauge between the mechanical extruder and the hotend. This arrangement can measure the back force on the filament, both detecting blocked nozzle conditions and unusual conditions such as hard contact between the print head and the print. Having said that, the more typical design has the combined hotend and extruder hung on a single quarter bridge tongue, an arrangement that adds so much mass that it further limits the slow electronics of the HX711.
I admit also to some bias as I am happy with a single underbed piezo sensor for obtaining the Z reference position and a touch sensor for mesh probing the bed. But even this is not a firmly held belief and the crash of favorite technologies and ideas being kicked off their pedestals often makes my workshop a very noisy place.
I have asked this on this forum, where most of the few remaining denizens are polite. Asking the same on Reddit or the Klipper forum would have my throat ripped out (figuratively) and my body trampled into the dust (figuratively)
So the Why?? remains. Is it perhaps that the Duet Smart Effector and now the Prusa Mk4 have started an avalanche of copycats? Maybe people found the choice of BLTouch, inductive probe, capacitative probe, and nothing else a bit stifling.

Re: Why strain gauge?
May 18, 2023 05:00PM
Your question is probably not answerable with any degree of certainty given that you are asking for an explanation of the actions of many individuals and companies. However, there are a couple of advantages that I can see for extruder-mounted strain gauges over piezos.

Piezos are sensitive to both deflection and rate of change, whereas a strain gauge doesn't care about the rate of change (as far as I know). That means that the time-course of the signal from a strain gauge is more easily interpreted than that from a piezo. Prusa seem to be working on the use of strain gauge signal for detection of extrusion issues, and that might be very difficult with a piezo.

Are the electronics for a piezo as simple as those for a strain gauge? The latter would need only a Wheatstone bridge (which is resistors) and an op-amp.

Also, I've read somewhere that "Strain gauges can be stunningly sensitive, fast responding, and mechanically stable for long periods - and they can be cheap as well." Maybe those things are relevant... ;-)

For the forces required to trigger the strain gauge, it seems that 500g is more than needed if the device is "stunningly sensitive". Maybe the implementations on Enders and the like are not well thought-out. I'm expecting that the Prusa Mk4 will use a more gentle tap than that.
Re: Why strain gauge?
May 19, 2023 04:32AM
@MJLew. It is true that there is no definite answer as to why strain gauges have suddenly become popular, I have wondered if I missed some breakthrough discovery or invention. I have experimented with almost all of the sensor methods that I have heard about and most have some advantages or other, but all have disadvantages or weaknesses.

Your point about piezo sensors measuring the rate of change of force has been bought up by many, but in the case of nozzle contact detection it is pretty much irrelevant as the nozzle contact event is dynamic - that is, it involves a rate of change. Detection of nozzle collision with the print is also dynamic, but the detection of extrusion issues that you mentioned, along with active pressure advance, both need a static sensor.

As far as the electronics go, piezos can even dispense with the op-amp and interface directly to an input pin on an MCU. Strain gauges can use a simple conditioning circuit, but most I have seen use a conditioner/ADC developed for digital weighing such as bathroom and kitchen scales.

On my printers, I use a conditioning circuit with a small MCU as this allows more than just nozzle contact detection - the software in the MCU can also measure the quality of contact, such as plastic on the nozzle.

I am presently investigating if it would be possible to use a strain gauge for detection of nozzle blockage, measuring back pressure for active pressure advance, and also as a fallback in case the underbed piezo fails to register a contact. The electronics for this may be possible with a single MCU reading the underbed piezo, touch sensor (BLTouch), and strain gauge.

Sorry, only registered users may post in this forum.

Click here to login