Duet setup - instructions in progress

[Eaglezsoar]

That would be Orange Menace, Eagle.

g.

My bad, menace it is.

[geneb]

So here's the question Gene, how many Max's do you have, if two are scratch monkey's? And what have you named them all? I'm more than slightly curious.

Orange Menace - v1 (Started out as an Indiegogo machine, got rebuilt with an updated Melamine kit to become OM. This is the machine in my avatar)
This machine was the genesis of the 1st Ed. v1 manual, some time after The Great Glasses Throwing Incident of 2012.

Blue MAX - v1 - Blue Melamine, yellow towers. (http://www.geneb.org/images/blue-max-cal-object.jpg)

Red Sonja - v1 - http://www.geneb.org/rostock-max/red-sonja.jpg

Orion Lannister - Orion (no pic) [GoT fans will understand why this is funny. ]
1st Ed. of the Orion User's Guide and follow-on updates

Casper The Friendly Rostock - v2 (sold)
1st Ed. of the v2 manual

White MAX - v2
2nd & 3rd Ed. of the v2 manual

Walter White MAX - v2
4th Ed. v2 Manual/1st Ed. v2 User Manual

I didn't paint the last three due to time constraints. At some point I may tear one of the v2s down and have the towers sent out for anodizing and do a crazy paint job on the Melamine. My next door neighbor scratch builds hotrods for a living so he's got some awesome paints available to him.

I've also got a LulzBot TAZ named Giggles that I have a love/hate relationship with. I screwed with it and now can't calibrate it. I need to man up and replace the damn Z carriages, but just haven't had the time. (I installed "anti-backlash" carriages in order to leverage a bed probe mod and it only worked a couple of times before telling me to See Figure One for future uses...)

I try to theme the names when I can. I admit to phoning it in on White MAX.

Walter White MAX came about when I was pondering the name - something to go with "White" and all I could hear in the back of my head was "SAY MY NAME!" *laughs* Walter it is. (Heisenberg MAX wouldn't have made any sense. )

You gotta have fun with these things - yes, they're a serious tool but you have to enjoy them.

My wife is a fan of purple, so I suspect it won't be too much longer before Pauline the Perilous Purple Printer of Prolonged Puns comes to be.

g.

[Eaglezsoar]

Thanks for all the names.
Pauline the Perilous Purple Printer is a great name.

[Windshadow]

I have always named my computer printers, starting in the spring of 1985 when I named my nice new Apple Laserwriter "Caxton" they all got names of famous early printers such as William Caxton the first to have a printing press in England and Franklin was another one.
My most current Epson is named after Caxton's successor Wynkyn_de_Worde but when I noticed that he has the same name as the founder of the first Discworld newspaper in the book The Truth https://en.wikipedia.org/wiki/The_Truth_(novel) on the Discworld created by the immortal Sir Terry Pratchett over some 40 plus books
he is my favorite author https://en.wikipedia.org/wiki/Terry_Pra ... rld_series so I decided to name my MaxV2 aft the the chief wizard of unseen university Mustrum Ridcully ...

so it is named Mustrum after him https://en.wikipedia.org/wiki/Unseen_Un ... chancellor and if I ever have this hobby to the extent it has infected Gene I have lots of other wizards of Unseen University to provide names for them.

Oh yes my avatar image to the right their is clipped out of a group painting of the wizards of unseen university.... the rather cheerfully mad and hooked on dried frog pills known chiefly by his title rather than his name, the Bursar. you can read his wikipedia entry right under the one for Mustrum in the link above

[geneb]

I've got a sneaking suspicion that your avatar is really Rincewind.

g.

[Eaglezsoar]

I've got a sneaking suspicion that your avatar is really Rincewind.

g.

Or perhaps SmellTheWind

[Windshadow]

Nope Rincewind is relatively young and agile and so he is very fast at running away it is main is his main survival skill, while the Bursar is far older (and befuddled most of the time) a much better fit to me

[Eaglezsoar]

Nope Rincewind is relatively young and agile and so he is very fast at running away it is main is his main survival skill, while the Bursar is far older (and befuddled most of the time) a much better fit to me

I can tell that you can take a joke and you are one of the users I would like to meet in person because of your great attitude and personality.

[mhackney]

Ahem, Duet?

[Rjaallen]

Hi mhackney,

A quick question, do you find that you need to use the H parameter adjustments when using your FSR mounts on the RMax? or do you get good repeatability across all 13 points despite only having 3 FSRs,

I'm noticing a skew in the printing plane but haven't had much luck removing it using the H parameter and following the steps of G92 Z0, Z+5, G30 S-1.

Also do you ever adjust the sensitivity jumpers on the JohnSL?

Cheers,

Rob

[mhackney]

Great question Rjaallen!

On my printers that have snow flakes (RMax) I do not need to tweak H. The rigidity of the snowflake makes the difference. In my printers with just borosilicate, an aluminum dissipator and silicone or kapton heater I find that the bed flexes ever so slightly between mount points more than over mount points, so I do tweak H.

I wouldn't think a skew would be related to that. But a quick check is to test Z=0 at each of your probing points to see if and how far off it is.

Once I have the printer setup up I tweak the jumpers to use the most sensitive setting possible. This has a lot to do with the mass of the heated bed, probing speed, mounting system, etc. On all my printers I am able to use the most sensitive setting with no problem.

Just an aside:

The thing folks don't think about or maybe realize - FSR measure force not mass. The equation for force is:

force = mass * acceleration

So, the sensitivity for the FSR is a function of how much mass is sitting on it and the acceleration at the point of contact/impact. If you probe too slowly (really really slowly) you could actually not trigger the FSRs at all. You can test this with a fingertip - push gently over a sensor and slowly increase the pressure. You should be able to push slowly and hard without triggering. FSRs do not measure pressure.

Cheers,
Michael

[Rjaallen]

Thanks for the reply,

This is exactly what I thought, It appears I may have a problem where the FSR is not triggering quickly enough due to the slow approach during the G32 execution in some areas,

i.e. the washer compresses a little before triggering FSR but not consistently across the bed.

When printing there is plenty of time for the washer to decompress. Possibly only on one FSR?

How thick are the silicone pads you are using? I used the thick ones from the ultibots kit, but I think this might have been a bad idea.

Perhaps I should try to increase the probing speed to see if this helps...

Oh and I have the snow flake installed, btw my reason for thinking along these lines is I've had some success by using false H values based on visual data during printing (too close to the bed as effector approaches z tower). However the values obtained by measuring H values do nut correspond with what I would expect due to visual interpretation of the bed plane during printing.

Thanks again,

Rob

[mhackney]

The compressibility of the pad you use is also a factor. The silicone pads that UltiBots uses are just right. Either the thick or the thin are fine.

You only need 1 FSR to trigger.

Try increasing the speeds in small amounts and test.

[Rjaallen]

Hi Michael,

Thanks for the help, I've got it working much better now,

It turns out that roughly doubling the probe speed helps a lot in my case, I used these parameters for M558 in the SD card config file. M558 P4 X0 Y0 Z0 F180 H3

My thinking is this, the FSR sensor measures Force but the JohnSL Board is constantly setting this back to zero. A bit like continually pressing tare on some digital scales

Consequently the probe measures Force per Unit time. And to trigger the probe requires a certain amount of Force to be added before the JohnSL resets in a certain amount of time.

Hence increasing probe speed decreases the time period for the force change to be detected and consequently increases the sensitivity of the system.

Anyway to summarise, I think this helps on my system as there is some non uniform x-y hysteresis effect occurring on my bed (or more likely in my effector/extruder, I'm using a modded flex3drive system with a lot of printed parts, and consequently the effector layout is not symetrical hence why I suspect this may be the root of x-y deviations) where the bed returns to 0 in some points a few seconds after probing hence the paper test didn't help me. Or in other words the slow probe speed allows the bed to be squished a bit before returning enough resistive Force to trigger the JohnSl board. By the time you start the print this squishiness has relaxed and the bed is now slightly higher in some areas than what was measure.

Thanks again,

Rob

[Eaglezsoar]

Ahem, Duet?

We do tend to get off topic once in a while, we will try to do better in the future!

[geneb]

MHackney, I see in your photo in post #1 that you've got some kind of crimp-on connector on your power & heated bed wires - what is it and where can it be had?

tnx!

g.

[geneb]

BTW, I've also found that the connectors that come on the steppers (black shell) seems to fit without problems on the Molex connectors used on the Duet. Is there any compelling reason to make the adapter or completely re-connector the cables?

tnx.

g.

[bot]

I find the pins work interchangeably between both connectors with a snug fit. I use both on my duet.

[geneb]

I've got my Duet hooked up to USB and I was able to successfully upload the 1.09r firmware to it.

One of the steps that the how-to mentions is using M115 to discover the current version of the firmware by using the Arduino IDE's serial terminal. At no baudrate did I get a response, before or after the firmware update. The computer clearly sees the port and it's correctly configured in the Arduino IDE (board & port selection).

I edited the config file in order to use the IP address I wanted as well as the other configuration changes recommended by MHackney.

The board appears to come up and I get the amber connectivity light and the green traffic light blinks periodically. However, the board doesn't reply to pings or web traffic. Running "arp -a" doesn't show the board's ip or mac address either.

Is 12V required for the board to boot completely?

Here's the config.g file I'm using: https://bpaste.net/show/50fda6d35397

If the lack of 12v isn't the problem, I'm not sure what to try next.

tnx.

g.

[bot]

I've never tried the arduino serial interface, but what version of the duet web interface are you running? Make sure you've uploaded the latest version to the /www folder on the SD card, and try the web interface again. It shouldn't require 12v, I don't think, unless your USB isn't providing enough 5v.

[geneb]

It never gets to the point of kicking off the server - the Duet doesn't even appear on my network.

g.

[IMBoring25]

Is the IP in your config.g a valid and available IP on your network?

[geneb]

Yes.

g.

[Lightpath]

Thanks for all of your help Michael. I have had a measure of success. It took some experimentation, and a lot of reading, but at this point everything is working well. I had some issues with the auto calibration, and had to play with my sensor z height a fair bit, but I got everything to work. Turns out my fsr mounts are fine, the neoprene washers did the trick- They work perfectly now.

Before I was running everything off of 24v, and it was working "ok". I decided in the end to go back to a dual voltage system. 12v for the duet and the hot end, 24v for the bed. It works well this way, and I had a couple of small 12v power supplies lying around.

Next task is to print your duet/Paneldue housing!

The only weirdness I have is that I can't auto-calibrate, print, then print again. It will print 3-5mm above the bed, extruding into air. I have to autocalibrate every time. Now since this takes all of a minute that's no big deal.

Thanks again, your coaching made this a lot easier!

[mhackney]

MHackney, I see in your photo in post #1 that you've got some kind of crimp-on connector on your power & heated bed wires - what is it and where can it be had?

tnx!

g.

Gene, these are Anderson Powerpoles and are perfect for this application. These are the 30A version and they come in 15 to 45 A. Bought in quantity that are affordable. The best solution for high current applications like this.