Conductive build plate?

Is there a way to make a polymer build plate which is also electrically conductive?

The reason I ask is that we're spending a lot of time making z-probes using nozzle contact and the gold standard we reference them to is direct electrical contact, which gives very accurate and repeatable results. We just can't print as well onto bare aluminium as we can onto various polymers.

Yes there are arguments against it but putting those aside, is there a way to make say PEI conductive? It doesn't need to be very conductive just enough to register when a nozzle at 5v/3.3v touches it?

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Ive been playing around with the idea of getting a thin sheet of metal the same size of my bed. The idea is to clap this down run the auto tramming then remove the sheet. If the thickness of the metal is right then you would be set at a perfect first layer height.

Such a build plate would be expensive? You'd need some inbuild collector and a connector.
I wonder if one could generate static electricity instead. ( my LokBuild plate is charged after a print and a contact-probe would have to deal with these surges, so why not use it as contact detection? )

Edited 1 time(s). Last edit at 07/13/2017 01:51AM by o_lampe.

Hi Scotty,

I'm not sure what you mean. If the metal is going to follow the contour of your bed exactly, what do you gain from placing it then probing? Surely you can probe the bed and add a small offset? Ahh I see what you mean, just as a probing tool? Hmm might work presuming it follows the contour of the bed, maybe just aluminium foil?

O_lampe, yes it might be expensive, but as long as its roughly the equivalent of a accurate z-probe such as our piezo20 combined with a decent build plate such as PEI then we have an integrated solution of reasonable cost.

Static or some other way of energising the build plate for a nozzle to detect (or vice versa) is also well worth thinking about.

Edited 1 time(s). Last edit at 07/13/2017 04:03AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

I don't see why you can't add metal filings to your polymer. The problem will be evenly distributing them to ensure the same conductivity everywhere on the build plate though.

It also means that you can only probe with a very clean nozzle. Any filament oozing out will stop conductivity.

So, process would be to heat up, clean the nozzle completely, cool down so no oozing is happening and then measure?

No we already deal with this with Piezo probe, heat your nozzle to 130 deg C and snip any excess material off with wire cutters. At that temperature even PLA doesn't ooze through a 0.4mm nozzle, and any residual plastic will squash on contact. Same technique would apply to electrical contact probing.

Or implement an automatic nozzle cleaning station.

Edited 1 time(s). Last edit at 07/13/2017 05:39AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
Origamib
I don't see why you can't add metal filings to your polymer. The problem will be evenly distributing them to ensure the same conductivity everywhere on the build plate though.

I wondering whether it would make a difference, as long as you get a signal when the nozzle touches the surface. You could amplify or condition the signal in some way, we have some experience with this, to ensure even a very weak rising edge would be considered a trigger. You can't get false triggers with this technique (unless from EMI) unless your using high voltage and expect arcing, its whether there would b a larger deviation of probing points between differently conductive areas.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

I'm all for mechanical precision, but all this worrying over exact measurement of nozzle to bed distance seems excessive. The bed surface isn't going to be perfectly flat, so your measurement at one point doesn't guaranteed the same nozzle to bed spacing at any other point. OK, you're going to measure at 10 or 20 places and use the measurements to map the bed for autoleveling. That will get it as close to level as possible, but it doesn't change the fact that the print surface isn't perfectly flat (and your guide rails sag, bearings wobble, etc.). Prints will stick to the bed over a relatively wide range of nozzle to bed distances.

If you're worried about exact part size in Z, a few 10s of microns isn't going to matter in any application for which a printed plastic part is acceptable, and again, will be dependent on the flatness of the bed and where you are making the measurements. A tiny bit of over or under extrusion in the top layer, caused by poor extruder calibration or variation in filament diameter, will guarantee that the absolute best you can do in printed part Z height is +/- several 10s of microns. Room temperature variation will cause the plastic part to shrink and swell by 10s of microns, so what temperature is the standard at which all your print measurements will be made?

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

I take all those points Mark, and I know generally you aren't a fan of software bed levelling etc... and neither am I when it comes to cartesians and corexys, it isnt necessary unless they are flimsy. My corexy is quite solid and once its mechanically level I use a microswitch z endstop and maybe babystep the first layer occasionally. I have one of my probes fitted to it for testing purposes only.

Perhaps use of grid levelling with tapering off is a valid reason to accurately map a non-flat bed and the more accurately you can do this the better.

When it comes to deltas that's a whole other ball game. Sure I can manually input the height differences at 16 points on my bed in Davids calculator and then update a config file and try again, but why should I when technology means I can hit calibrate, and match my printers mechanics to its bed, to a deviation across 300mm diameter of 0.006mm in a few seconds? Do I need it to be this accurate, no but then did 18th century histologists need electron microscopes?

Simon.

Edited 2 time(s). Last edit at 07/13/2017 10:45AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Sorry to not answer your direct question; we recommend the nozzle be at 130 deg C when using piezo probes, with bed at first layer temperature in case it warps when heated. This is to standardise the procedure as best we can, to ensure any residual filament on the nozzle is soft and therefore doesn't give a false reading, and because this is a temperature where no ooze should take place with common filaments.

As for an electrically conductive build plate, you would probe with bed at first layer temperature with a nozzle at 130 deg C for the reasons as stated previously. You need a clean nozzle, but seriously are people really printing with a nozzle covered in blackened, burnt plastic because cleaning it is such a drag? As an argument against direct electrical contact as a probing method this is a weak, lazy argument.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
DjDemonD
As an argument against direct electrical contact as a probing method this is a weak, lazy argument.

I'm not so sure. I assume most high end printers produce enough force to snap a heat break, so probing the bed with an unreliable sensor seems like a high risk play.

And your example of a reliable sensor?

Probe with 50% normal motor current if you're worried. I've tested piezo modules through tens of head crashes and haven't snapped a heatbreak yet. It's only probing at 5mm/s.

Edited 1 time(s). Last edit at 07/14/2017 09:43AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

LNP STAT-KON EE004 is a compound based on Polyetherimide resin containing 20% Carbon Fiber. Added features of this material include: Electrically Conductive.

[www.sabic-ip.com]

PS.: you can make your own: electrically conductive PEI pellets.

Edited 1 time(s). Last edit at 07/14/2017 10:10AM by newbob.

Now that does look interesting. Thanks.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
DjDemonD
And your example of a reliable sensor?

Probe with 50% normal motor current if you're worried. I've tested piezo modules through tens of head crashes and haven't snapped a heatbreak yet. It's only probing at 5mm/s.

To be fair, The first time I hooked up an IR probe I accidentally had a nozzle collision and left a nice dent. These days I hover over the emergency stop button, but regardless of speed you might have a serious nozzle collision and it is always worth considering back up. When the nozzle is touching the bed, there is zero room for error and all the most expensive and important bits of your printer are coming into contact. At least with an end stop you are slightly above the bed. Hence why I asked in the piezo thread about back ups. The IR sensor can have an end stop as well to stop collisions in case of fault.

When I eventually hook up a nozzle contact sensor (the IR probe is not playing nicely) I will also be using an endstop 50~ microns below the bed surface in case of failure to stop serious damage and will also run the motors at 50% (Good idea!). It's always a good idea to think of potential points of failure, but I do not see why any of these should stop a conductive sensor on the hotend.

Having an endstop as a backup, you can connect an unlimited number in series with RRF and set them as limit switches to run a macro if triggered, they're all in NC configuration so if any fail you fail safe, positioned just behind your regular endstops, makes sense. Low motor current makes too.

Edited 1 time(s). Last edit at 07/14/2017 12:59PM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

I spoke with one company who can supply a 9.5mm thick sheet of conductive ultem/PEI 300x600mm for $970.

So might be an expensive surface to work with even if I could slice it into 4x 300x300mm sheets and unless adhesion without heat was very good it might be useless used over a heated bed.

Edited 1 time(s). Last edit at 07/18/2017 09:05AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Maybe something like this?: Conductive PC
That is considered conductive, but sounds expensive, even without a price given.
There are other options like that. Just search for "ESD safe polycarbonate", or whatever material you prefer.

This ESD safe PC might only have a coating of a conductive spray. After you've sanded the surface, the coating is gone.

Yes these materials are either quite limited or very expensive. But I'll keep looking.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

DjDemonD,

Look at polyanilines ( PANI ), and its family of conductive polymers. Been around longer than most plastics.
They're used in making transducers, like photovoltaic cells, though not very rigid nor considered really flexible.
Problem is it's usually an OEM material, and you'll need to synthesize it yourself, as sources are rare.
Depending on the the base reagent, various colors can be formulated. I'll leave that up to you to research.
Might make a great build plate, or not. No telling how "sticky" it will be for different filaments.
Be cool if it could be cast directly on aluminum, like polyurethane.