Simple Metal Freezes about five to ten layers in

[Jack Crow]

Hi all,
Have a simple metal with some minor updates. New board, new hot end, but in essence stock.

Since coming to Kansas it has developed a new problem.

Starts and runs normally and after about five or six layers, it simply stops.
It keeps regulating the heated bed properly.
Im assuming the hot end is doing the same.
Motion stops.

The section on Cura that give the status report starts off with this...
< Bed x: 20.00 y: 142.00 z: 0.13
< Bed x: 20.00 y: 10.00 z: 0.64
< Bed x: 142.00 y: 10.00 z: 0.62
< echo:endstops hit: Z:0.62

when the print starts it shows the above, and when it stops nothing new is added.
If the USB cable is un plugged during a run, I get a message about that. (don't use the memory wafer feature)

USB cable seems good, as do the sockets.
Fairly sure about the power supply.
Suppose I can hang a scope on it and have it record drop outs.

Built a 'meter' house that watches unit voltage and current draw, those readings are normal.
12 volts up to 7 amps or so when everything is running.
Been that way since I put the meter on.
Don't think that is the issue.

Did some reading here.
Seems thermister issues can do this.
Can hang a scope probe on them as well if I have to.

In other reading it could be a firmware bug.
With PBHQ off line, I have no idea where to get that from or how to re flash the printer. (More high voltage analog than a digital dude.)

Gave the flexible wires a tug,
Had to replace some of those in the past when they fatigue out.
The unit stayed working.

Open to ideas and things to test.

Thanks in advance.
Jack Crow now in Kansas.

[RetireeJay]

Jack,
I'd highly recommend using Repetier for diagnosing this. It will show you all the traffic going back and forth between the computer and the printer (even keeping a log), and it also shows you the current X, Y, Z coordinates, bed temperature, hot end temperature, and fan percentage. And best of all, you can have a graph of the bed & hot end temperatures vs time, so if anything flaky is going on, you have a pretty good chance of seeing it on at least one of those "channels".

Repetier does ask for a donation, but you can download without donating. If it solves your problem, maybe it's worth the $10 they are asking.

Offhand, I'd suspect a problem with a fatigued wire, probably a thermistor wire, but possibly a heater wire. Once a print starts, the printer is "blind" to the "endstops" and it also pays no attention to stepper motors stalling out or anything like that. If it can go through the steps of calibrating X, Y, and Z and starting a print, then there's no reason for the firmware to stop running unless a temperature has gone out of range or (much less likely) there's a problem communicating with the computer over the USB cable (which could be in the computer or in the cable). Re-flashing the firmware should not even be considered until all other options are exhausted, and this symptom does NOT point at a corrupted firmware image.

Wow, moving from VA beach to Kansas... from ocean front to as far away from salt water as you can get in the US. Trading hurricanes for tornadoes. Hope it was a very worthwhile move for you!
-RJ

[novice]

...In other reading it could be a firmware bug.
With PBHQ off line, I have no idea where to get that from or how to re flash the printer. (More high voltage analog than a digital dude.)...

If at some point you have the need and wish to reflash the board, the resources necessary are all linked in this thread http://www.printrbottalk.com/forum/viewtopic.php?f=15&t=11644 including a still active tutorial.
Trust me when I say if I can do it, you can too.

[Jack Crow]

RJ
Answers interlaced in your note...

I'd highly recommend using Repetier for diagnosing this. It will show you all the traffic going back and forth between the computer and the printer (even keeping a log), and it also shows you the current X, Y, Z coordinates, bed temperature, hot end temperature, and fan percentage. And best of all, you can have a graph of the bed & hot end temperatures vs time, so if anything flaky is going on, you have a pretty good chance of seeing it on at least one of those "channels".

(Ok.
Did the down load thing.
Skiped the option of paying for it since Im still broke.
Have no idea what to do with it.
Lots of settings and not much of it makes any kind of sense.)

Repetier does ask for a donation, but you can download without donating. If it solves your problem, maybe it's worth the $10 they are asking.

(Back to the question. What is it and what will it do. Mostly how to use it.)

Offhand, I'd suspect a problem with a fatigued wire, probably a thermistor wire, but possibly a heater wire.
Once a print starts, the printer is "blind" to the "endstops" and it also pays no attention to stepper motors stalling out or anything like that.
If it can go through the steps of calibrating X, Y, and Z and starting a print, then there's no reason for the firmware to stop running unless a temperature has gone out of range or (much less likely) there's a problem communicating with the computer over the USB cable (which could be in the computer or in the cable). Re-flashing the firmware should not even be considered until all other options are exhausted, and this symptom does NOT point at a corrupted firmware image.

Wow, moving from VA beach to Kansas... from ocean front to as far away from salt water as you can get in the US. Trading hurricanes for tornadoes. Hope it was a very worthwhile move for you!

(The problem was VB is a Navy town, and most anything the Navy wants done requires a clearance. Had one for years but it has since expired. Now thanks to that idiot Snowden and friends, it takes over a year to get it re instated. So Norfolk / Virginia Beach is a loosing game. Got a call from a head hunter for this gig here in Kansas. If all goes well it will pay very nice, for the next four months or so. It was supposed to be for six months but the company says it ends on 31 December. Still an improvement.

Im going to check out Youtube and see if there is anything on this program.

Thanks in advance.
Jack Crow

[Jack Crow]

RJ
Just tried to use cura and it can't find the printer any more.
Clearly I am missing a clue.

Jack Crow

[RetireeJay]

My workflow when using Repetier is to FIRST slice the object, THEN use Repetier Host to print it. You can use your existing Cura to do the slicing, just save the G-code and then load the G-code (not the STL) into Repetier. That bypasses all the settings, which have to do with setting up the slicing programs supplied with Repetier (namely Slic3r, Cura Engine, and Skeinforge).

The only setting you need to worry about in using Repetier Host as your diagnostic tool is to connect it to the correct COM Port. It may be able to find the port on its own, but if it doesn't you can find out the correct port by looking in Device Manager. (Oh, and MAYBE you'll have to play around with the Baud Rate, but probably not.)

Umm... another thought just hit me. Repetier sells at least three products: Repetier Host, Repetier Server, and Repetier firmware. You want ONLY Repetier Host. The Server just makes things more complicated unless you really need to set up a wireless connection and a Raspberry Pi between your computer and the printer. Server does have a bunch of settings; I took one look at that and ran in the other direction fast!

[Jack Crow]

RJ and team,

Got some follow on information.
That Repetier program got me so confused, it totally stopped up the works and gave exactly zero in terms of satisfaction and answers.
Will need hands on help to sort that mess out.
Today will not be the day.
Monday ain't looking good either.

Managed to un install the program and all went back to normal.

To make maters all the more complex, I am living in a hotel room in Lenexa Kansas and have very limited resources.
But I did bring a scope with me. I have slightly different priorities than most people.

Below is a link to a photo article on some of the trouble shooting.
https://imgur.com/gallery/6hSyKZl

It boiled down when the scope showed a positive voltage excursion, more often than not, the printer would freeze.
The guess is the micro processor has a voltage sensor in it, and shuts things down when tripped.

Did not have my collection of caps but did have a gell cell battery. In a pinch a battery makes a fine cap.
For the first time in a long time was able to get about 70% of the way through a print before the freeze.

Need to see if a PSU can be found in Kansas on a Sunday.
Going to visit the local Micro Center and talk up the issue, perhaps someone has a Meanwell PSU on the shelf for sale.

Will report back one way or another.

Jack Crow

[novice]

I too have had program conflict issue with past attempt to use Repetier-Host 1.0.5 and Cura 15.04.6 on the same computer. Had figured it was because I use underpowered or obsolete machines, now I wonder.

Just thought I'd mention for anyone reading this, after performing an uninstall on Cura (uncertain about Repetier-Host) a folder will remain on the hard drive under >users>applications>cura that should also be removed.

[RetireeJay]

Funny. I have both Repetier and Cura installed. I tried out Cura once, and went back to Repetier, but never had a conflict (even with both programs running at the same time, just not "connected" to the printer at the same time).

[novice]

Funny. I have both Repetier and Cura installed. I tried out Cura once, and went back to Repetier, but never had a conflict (even with both programs running at the same time, just not "connected" to the printer at the same time).

I'm guessing you're running something more powerful & recent than a Win 7 machine with a Celeron N2840, 4GB ram ??

[Jack Crow]

RJ
Had all kinds of trouble yesterday between the flaky hotel room internet and the web site.

Here is the news.

Have a modified server PSU that I got at a ham radio event. 12 v 26 amps, think I paid 20 bucks for it.
It was designed to be hot swap able, in a rack. Had to make a cover to connect the I/O’s and hold a fan.
Made a real poor choice, that fan makes a racket.

Anyhow, did another scope reading and the ripple was less. Added a loose cap I found in my kit and that took off a lot of the hash. Not perfect but real well.

1. The upshot of all this. Finished my first print in about 2 weeks. Then almost finished my second print, ran out of plastic! The nature of luck.

Here is what I suspect, when the bed heater is turned off, there is an upward spike in the voltage, couple that with the noise the steppers make, if two release at the same time, it trips an over voltage detector in the Printerbots micro.

Could not find a proper power supply in Kansas on a Sunday. Did an Amazon order for a 12 volt 30 amp. This one is adjustable around a 12 volt center.

To get this all to work, had to do some extensive wire work.

Will try and post some photos.

Keep it safe
Jack Crow

[RetireeJay]

Funny. I have both Repetier and Cura installed. I tried out Cura once, and went back to Repetier, but never had a conflict (even with both programs running at the same time, just not "connected" to the printer at the same time).

I'm guessing you're running something more powerful & recent than a Win 7 machine with a Celeron N2840, 4GB ram ??

Actually, I am running a Win 7 machine - albeit with 8G ram.

[Jack Crow]

Hi all,
Think of this as an update.

Got the new power supply in from Amazon.
Did the wire work, spot checked it with the meter then connected it to the Printerbot.

First print failed by freezing after about 5min.

Added 1500uF of cap to the output of the supply.
Had a successful 1 hr print.

Now working on a 4 hour print.
A few more of these and I can say the issue is solved.

Set the scope to 5s per CM and watch the maximum excursions.
Seeing spikes that cover a p to p of about 115mV.
Also set the new PSU to about 11 volts to have a touch more headroom on the positive excursions that seem to be causing so much trouble.

The negative spikes are caused by the bed heater turning on, sucking down 7 amps or so.
The positive spikes show up when the bed heater is shut off.

The hot end can't draw current like the bed heater and hardly shows up.

As for the external cap issue.

Beginning to wonder if the board has it's own built in filtering that gave up.
If so, next time I have it out, will change out those parts for Low ESR types.

Also thinking about locating some parts to make a hash filter.
Can get most of what I need from a dead PC supply.
Not ready to sacrifice the unit I got from Printerbot just yet.

So that's the news from Kansas.
I do like it here. Good people. New things to learn and try.

No cyclonic storms yet. Perhaps I will get the chance to see one.

Jack Crow

[Jack Crow]

Hi all
More follow ups.

I had just about given up on this printer.

Since I seem to have a lot of time on my hands took one more lunge at it.
Ran the printer at 10 volts.
No improvement.
Ran it at 14 volts.
No change, still exhibited random freezes.

One of the last things this priinter did was drive the print head into the work surface.

Digging a hole in the kapton tape and yet another gouge into the aluminum work surface.

Did some test prints and noticed that the molten plastic was not extruding properly.

Machines will do dumb things from time to time, and trying to push the print head into the base is one of them.

Had some spare nozzles.
Changed out the .4mm for a .6mm (was fresh out of .4's).
Suddenly things started to work better.
Have made a couple of four hour prints, and no freezes.

Ran it all weekend and no flaws.

Wondering what I am missing?

The circuit board power connector and it's mate is showing no signs of heat stress or excessive oxidization.

Right now it's running at 14 volts, the bed heater works faster.
It's behaving properly. Grateful for the result but I don't understand it.

Keep it sane.
Jack Crow

[Mooselake]

A mostly random guess, but wonder if the increased load on the extruder stepper made a difference? Maybe caused a problem with the stepper driver chip that somehow affected the processor? It's still pretty odd.

Clean out the old nozzle with a welding trip cleaner, the IPA burnout trick, a wire drill (careful, don't enlarge it) or your other favorite method. I'd also suggest extruding a few mm of plastic with the nozzle removed, might flush out some crud. It's generic advice, but seems to help on a regular basis...

Might be too late for regular sanity
Kirk

[Jack Crow]

Kirk
Had a first freeze in a week.
Seems the whole computer crashed. The printer was not a fault.
When everything was re booted, once again Cura went stupid and forgot all it's settings.
Nothing to do with the printer this time. This is an old problem.

Otherwise I am back making parts.

Don't know how failure you are with RF lab stuff.
The current day job we make use of directional couplers, attenuators, and power meter sensor heads, to make transmitter measurements.

In the past people use blocks of wood, books, office supply's all kinds of little crap.
Deluxe shops have these little steel mechanical jacks.
The idea is to hold all these things to keep the parts off the bench and at a consistent height.
This takes stress off the connectors.

If these experiments work out, it might be the beginnings of a product line. I like the sound of that.

Been using the printer to make holders for different combinations of attenuators and couplers.
Some are larger than others. Worked out two standards.

The bigger ones have a 55mm center of the RF connector to the base. Those units are like a big Fosters beer can size and very heavy.
Right now working on ones for these Aeroflex attenuators. Longer and skinnier, so the second standard will be a 30mm center to table top.

Most guys in two way tend to think in terms of CW power.
The rest of the world works in Pulse Power. So these attenuators don't usually get too hot to handle for bench work.

That's where the head is at. Will get permission from the day job to bring in a camera and take some photos.
Right now at this extended stay motel, I have a black table, black PLA parts, and a black attenuator.
Tough to photograph.

Currently working near Kansas City. It's a contract gig that times out around New Years.

More when I know more.
Jack Crow aka Radio Mike