Heat Bed default

Hi,
I keep going on trying to use my Ormerod 2. Trying to print ABS I started by printing the printer parts (and I was glad I saw this before printing other things because my nozzle mount failed pretty quickly. I was glad also to notice a subject mentioning the original parts on github, because I printed a few newer parts incompatible with my nozzle mount before noticing).

I fixed a problem with the heatbed: there was a bad smell coming from the connectors (first version of the Ormerod 2) and the left M2 screw broke when I tried to tighten it. By an incredible luck I found another one and soldered it in place. exposed PCB trace was damaged also so I soldered a tiny piece of wire (16A) on it to increase the conductivity. Cold sweat during a few moments.

While I was at it I upgraded the firmware to the latest DC42, and while messing around with the parts I made a mistake and dropped the duet case. Which was still plugged and opened, so some connectors where shaken a bit. Silly me, I know...
I think the thermistor connector is damaged, because sometimes the read temperature falls. I'll have to change it I suppose, but right now if it doesn't move the reading stays OK (I hope).

Now for the latest problem: I tried to print a piece in ABS, and the heat bed stopped heating after a layer. The temperature just drops as the webpage indicates "default" under the bed. It is an intermittent failure but once it occurs the firmware doesn't try again. The thermistor contact doesn't seem to be the root cause because the message is: "Error: heating fault on heater 0, temperature rising much more slowly than the expected 0.0°C/sec"
What in your opinion could be the problem?

Edited 1 time(s). Last edit at 09/12/2016 05:40AM by mat_fr.

It could be the thermistor connection, but more likely to be the bed connection failing. Check your repair, and also tighten all the terminal block screws on the Duet and 5V board (if fitted). The syptom of a failed bed connector is that the LED on the Bed does not light even though the bed LED on the Duet (D6) is lit.

New bed heaters are not all that expensive. This one looks pretty good: [www.amazon.co.uk]

Dave

Thanks for the reply Dave.

Terminal blocks are pretty tight.

Actually I saw the LED on the bed blink too quickly for a normal temperature regulation process (didn't think about looking at the Duet LED...) and then the LED went dim, barely visible.
Naturally I thought it was a short circuit. Once the power cut I checked the continuity and saw that there was a non null (few hundreds of ohms, but highly variable) path between the bed connectors and the metal enclosure. I don't recall this as supposed to happen so I'll have to double/triple/... check all contacts and hope that it's not that bad.

Thanks for the link also. Are these rubber heat beds easily installed on the Ormerod ? Sorry the question is naive but the stock ormerod2 is my only experience with 3D printer and I find it very delicate to handle so I've become very cautious about touching it (less and less, that being said )

Quote
mat_fr

Now for the latest problem: I tried to print a piece in ABS, and the heat bed stopped heating after a layer. The temperature just drops as the webpage indicates "default" under the bed. It is an intermittent failure but once it occurs the firmware doesn't try again. The thermistor contact doesn't seem to be the root cause because the message is: "Error: heating fault on heater 0, temperature rising much more slowly than the expected 0.0°C/sec"
What in your opinion could be the problem?

A few people have had similar problems using under-powered bed heaters with 1.15 firmware. Here are some suggestions:

- Tune the bed heater if you have not already done so. See [duet3d.com].

- if the problem is that after the first layer the print cooling fan comes on and the bed heater can no longer maintain the requested temperature, change your slicer parameters to use a lower bed temperature after the first layer

- if all else fails, revert to RRF 1.14 and DWC 1.11

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I hadn't done the tuning yet, I'm running it just now.
Is the PWM intended for the heat bed or do I have to set it to 1? I suppose PWM is a good solution to reach a stabilized temperature but won't the rise going to be slower than ideal?

I don't have a print cooling fan, just the hot end one, which is not controlled.

Edit: results are

Heater 0 model: gain 47.5, time constant 473.2, dead time 10.5, max PWM 1.00, in use: yes, mode: PID
Setpoint change: P169.9, I0.36, D1245.7
Load change: P169.9, I8.11, D1245.7

I can't evaluate these results
I'll try a print this afternoon and see how it goes.

Edited 2 time(s). Last edit at 09/13/2016 05:56AM by mat_fr.

That gain parameter looks low to me. What is the maximum temperature that your heated bed can reach?

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

It can reach 105 if I let him enough time (like 30mn or so, I never measured but it seems endless).
I often select 110 in the preset list by mistake and it won't take any other command until the temperature is reached. Which it basically doesn't. Dead end.

I can't say if this is new, I had never set it to more than 60 before trying ABS...

Quote
mat_fr
It can reach 105 if I let him enough time (like 30mn or so, I never measured but it seems endless)..

Better isolation under the bed will help a lot, my Ormerod 1 with the original Alpine ATX PSU (11.00V at Duet in) goes to 116.7C in 11 minutes




.. and 150.2 in 20 minutes, not recommended with this type of isolation though :-)



Erik

Quote
mat_fr
Thanks for the link also. Are these rubber heat beds easily installed on the Ormerod ? Sorry the question is naive but the stock ormerod2 is my only experience with 3D printer and I find it very delicate to handle so I've become very cautious about touching it (less and less, that being said )

Depends on your bed arrangement - most people have altered it from the original. You could perhaps keep the PCB heater as a support, but not connect it, and lay the silicone heater on top of it. But that link was only an example, you can also buy a PCB heater of the correct size for little cost, just search Amazon or eBay.

For ABS you really need a couple more volts fed to the bed heater to get to temperature in a reasonable time. You can buy 12V 30A LED power supplies that have a voltage adjustment (e.g. [www.amazon.co.uk] though there are cheaper ones available also, just search). You will need to print a suitable cover for the terminals, preferably with a mains input connector and maybe a 12V output connector (ensure the 12V connector is rated for high currents). I use one of these PSUs and have cranked the voltage up to 14V which gives me a bed heating time of 7 minutes or so to reach 110 deg C - less if I put a cardboard cover over the bed while it is heating.

Dave

My power supply has a potentiometer to set it to the correct value. I've read about pushing it to 13.5V but I'm afraid the linear regulator wouldn't appreciate it too much (it is indeed relatively hot on normal operation).
I can try this solution also.

OK so now I have a bed that seems to heat correctly, the temperature is rising and correctly measured (it seems), the Duet LED is ON, the heatbed LED is OFF (or so dim it's barely visible)... No print running, the arms stands at 10cm above.

Regarding temperature reading, does it seem possible to you that, with a bad contact on the connector (after the fall issue) the resistance is higher than expected and thus the temperature lower?

Oh, the heatbed LED is back on ^^

Edited 1 time(s). Last edit at 09/13/2016 08:02AM by mat_fr.

After changing the heatbed sensor connector and taking the duet out of its inclosure/adding some kapton tape and taking all back together (no more continuity between the case and the heatbed connectors, here are the new results:
Heater 0 model: gain 76.9, time constant 440.9, dead time 8.4, max PWM 1.00, in use: yes, mode: PID
Setpoint change: P121.8, I0.28, D716.5
Load change: P121.8, I7.25, D716.5

The gain is much higher, but the LED still misbehave, with 1.4V drop only, when the bed is powered by 11.56V (I forgot to boost the supply voltage a little)

It's normal for the LED to blink rapidly when running the bed in PID mode, or when tuning with a P factor below 1.0. it's not normal for the LED to go dim.

In principle, the gain is about equal to the highest temperature that the bed can reach less the ambient temperature.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Ok, I can understand that. Problem is, I run it with PWM set to 1 (didn't know if less was ok with heatbed) and when the LED goes dim it's pretty much OFF really, although the LED on the Duet is ON.
But sometimes the bed LED turns back ON at full brightness.

I increased the voltage to 13.5V, this time the tuning canceled because 110°C was reached (it was the limit I gave it). I don't know how high a temperature the test is supposed to go.

If you run at a higher voltage, probably best to remove the external 5V regulator and enable the Duet's on-board switchmode regulator.

Dave

Let me see... 68°C measured by IR thermometer on the surface of the 7805. Maybe I should print directly on the regulator, that would stick

OK, I might just enable the onboard switcher. No problem to be expected from the ripples?

I might have a problem with the wires too short to reach the Duet screw terminal directly though.

Edited 1 time(s). Last edit at 09/13/2016 12:49PM by mat_fr.

OK now my extruder stalls
Could the increase in voltage cause extruder malfunction? The controllers don't seem overheated, and the hot end temperature seems correct, though the fan runs faster. I imagine thermistor is read through a regulated 5V so increasing the power supply output should not change the reading...

I changed the spool to PLA to see if I could extrude or if the nozzle was really blocked (I didn't want to set the temperature too high) and indeed it flowed, but after some orders of extrude 5mm there were some stalls again.

On the plus side, the bed got to 105°C quicker than before and was indeed stable for the small duration of the test.

I unclogged the nozzle gently using a 0.5mm drill bit and the extrusion went OK. For a time.
The next print showed deformations on the Y axis (missed steps?) and in the middle of it power went out. The fuse is OK, the Duet is alive went connected by USB, but the power supply is out
No smell, no burnt part visible inside (I haven't opened it to see in detail), the capacitors where capped with a thin protection that went off (the glue melted perhaps) but the capacitors themselves have not vented out. It might have over heated (again it's the first version of the ormerod with a fanless supply).
I suppose I'll try to see what's really happend this week-end, ans probably buy a replacement...

Edited 1 time(s). Last edit at 09/14/2016 03:02AM by mat_fr.

I really miss Ian's interactions in these sort of posts. He had a knack of boiling it down to what a noob needed to know.

Like "how not to burn the power supply" ?

Which power supply is not working, your main 12V supply or the onboard 5V switching supply? If it is the main 12V supply all your symptoms make sense. I suspect that it began cutting out for brief periods before finally failing altogether, which is not an uncommon failure mode for a switching supply. The brief cut-outs would not be long enough to kill the 5V and reset the Duet, but would be long enough to cause missed steps. Perhaps the PSU did not like the increased power demand after you put the voltage up - probably the best course of action would be to buy a new 30A or 40A power supply from eBay or Amazon. Getting from China would be cheapest, but will take ages to arrive and will probably not be built to European safety standards, so better perhaps to pay the extra and get a CE certified PSU such as the one I linked to delivered in a day or two. You can get away with not having a cover on the supply until your printer is working so you can print one. Just be aware of any exposed live mains terminals and ensure that pets and children don't have access, and the fact that the mains lead will have no strain relief, so don't trip over it.

An IR thermometer will read OK on an I.C., but they do not give accurate readings on the aluminium hotend (area too small and reflective) or bed (material has poor IR emissivity). A thermocouple thermometer is best. For the hotend, unclip the Bowden tube from the extruder and feed a wire-leaded thermocouple into the Bowden tube until it bottoms in the nozzle and you will get very accurate readings.

I have found that if you leave residual plastic cooking in the hotend without extruding for longer than a minute or two, you'll get a blocked nozzle soon afterwards, so be quick with the temperature measurement and never leave the hotend at temperature for long before extruding. Using a drill to unblock will work, but it may have scratched the walls of the hole and/or liner which will reduce the quality of the extrusion and also left brass or PTFE debris that will cause another blockage soon, so maybe buy a few spare nozzles (with some PTFE tube for the liner) while you are at it.

All a PITA I know, but I assure you that it will be worth it in the end!

Dave

The main 12V power supply. The 5V switching power supply I haven't tested yet but I removed the linear one and switched the jumpers according to this subject (http://forums.reprap.org/read.php?340,705962) so I'll test it as soon as I have a new PSU.

I suppose it all makes sense, indeed...

I've ordered this one, mainly because it was deliverable tomorrow. Hope it will do: [www.amazon.fr] (french page but the product characteristics are clear). I just don't know how it will fit but I'll figure something out if it doesn't. I was beginning to consider using an old PC ATX power supply, plenty of power and readily available in my cupboard. I may test it the evening, just for the sake of experiment.

Funny thing is that since I got the printer back in service 2 weeks ago I had all problems possible (no, I should not say that, I know there can be plenty more) but managing to print beautiful pieces made me eager to print more, so my enthusiasm is still intact (really what is the most annoying with tinkering with the printer is the Duet enclosure/electrical connections. Managing to plug the 12V connectors into the screw terminals is nearly impossible).

Thank you all for your patience and advices. I hope I won't need more too soon.

Edited 2 time(s). Last edit at 09/14/2016 09:10AM by mat_fr.

Temporary setup.
Behold! I present to you the Ormerod 2 Cthulhu edition (aka "don't do this at home, it will drive your wife crazy"). It's ugly but it seems to validate the root of my troubles were indeed the power supply.

The heat bed LED is still unreliable, and the heat bed heats slower now as the ATX 12V is fixed, but apparently I have no problem right now.

I'm printing the extruder gears in ABS, I'll see if they fit.

At least you are running again! The bed LED issue suggests that you still have a bad connection between the Duet and the bed heater. If the heater PCB is damaged it is probably easiest to buy a new one from Amazon or eBay. Look at the dimensions, voltage and power carefully to ensure you get a suitable one.

The original Ormerod that I have was supplied with an ATX PSU. The 12V was not that good - most ATX PSUs don't stabilise the 12V too well, and while the total power of the PSU may be 300W or more, it expects most of the power to go to the lower voltages and the 12V can sag as low as 11V at high currents. Changing that PSU was the first mod I made.

I have not attempted to attach my PSU (which looks the same as the one you are getting) to the printer. My PSU sits on the floor in its printed case (keep the air slots clear!), and I have a high current connector on the PSU for easy detachment, and very thick wires to carry the 12V to the printer. I have toyed with the idea of bolting the Ormerod to a thick plywood base with "L" brackets and mounting the PSU (and maybe the Duet) onto that. This would help keep the printer mechanically stable, especially the front and rear Y rods that tend to move in relation to each other with the slightest force on the frame, which then makes it impossible to get the bed level. It could even have a plywood back wall to attach the Z extrusion to, which would keep that perfectly square, and later expanded to form a completely enclosed heated cabinet to stop ABS warping. But as I've mentioned before, I have instead decided to design and build a completely new printer from the ground up, so doubt I will make any further mods to the Ormerod.

Dave

In fact, the PSU I'm using right now is my former PC supply. It's given for 2x25A for 12V but it failed some time ago, the PC started but without display. I assume the 3.3V was to blame, but for the time being the 12V rail does the job. I just received the new one but I'll install it during the week-end I think, for now it's printing

I'll consider buying a new heat bed also.
Again, thank you all for your advices, it was really a great help.

Hi,

Sorry I'm back with a smaller issue (way smaller, prints are OK)

Quote
dmould
I have found that if you leave residual plastic cooking in the hotend without extruding for longer than a minute or two, you'll get a blocked nozzle soon afterwards, so be quick with the temperature measurement and never leave the hotend at temperature for long before extruding. Using a drill to unblock will work, but it may have scratched the walls of the hole and/or liner which will reduce the quality of the extrusion and also left brass or PTFE debris that will cause another blockage soon, so maybe buy a few spare nozzles (with some PTFE tube for the liner) while you are at it.

I'm printing a piece right now and I noticed a small defect which seems that this one : [www.simplify3d.com]



I changed the PSU, it works really better now, in terms of bed temperature and speed. I wonder if the real voltage compared with the preceding one could have an effect on the mechanical precision (on the motor side). Maybe also the Z runner is weakening, it's still the original PLA piece, I have not yet integrated some of the new ABS parts I've printed, and the z runner has a small crack.
But in the end I'm wondering if the nozzle is not to blame. In that case I will buy a replacement one (or a few). But having the first version of the ormerod I don't know which nozzle I should take, if there are incompatibilities with the hardware. I've seen I would have to print the nozzle mount, but other than that I don't know what to buy...
And would it be worth taking a smaller nozzle than the stock 0.5mm when I'm at it?

The symptom you show is almost certainly caused by a loose X belt. The Y side is not clear, but if that also have layer lines then the Y belt is also on the loose side. PSU issues would not cause that (a missed step will show a definite offset rather than wandering layers). A loose belt will result in the perimeter being laid down in a different place if it is drawn in one direction compared to being drawn in the opposite direction.

Other issues that will cause layering are an overly bent Z screw or a screw that is constrained at the top (it should be free to whip around at the top), and can also be caused by the filament coming off the reel in jerks, which pulls on the extruder and pulls the X-arm to one side with every jerk.

Dave

Thanks for the diagnostic.
The X belt seems tight enough. The Z runner was loose so the arm could move a little. I've fixed that. The Z axis rod is not absolutely straight though. It's not constrained at the top but I intended to print a Z Top piece to support a spool holder, which would include a hole for the Z screw. Maybe that's not such a good idea if I read you correctly.
And the spool is definitely not held as it should. I struggle to find a way to hold the spools correctly: the original solution only works for small spools and the ones I printed are not so effective in locking it in place while allowing a fluid movement.

No, definitely do not constrain the top of the Z screw. A quite badly bent threaded rod will still work fine if it is held in only 2 places (the lower gear and the nut-trap). The top of a bent rod will describe large circles at low Z heights and smaller circles at higher Z heights, but will still raise the X-arm by the correct amount without undue sideways forces. If it is constrained at the top however, you are forcing the middle of the rod to attempt to move around in circles rather than the top, and this will cause it to exert considerable sideways force on the nut-trap which changes in direction as the rod rotates (moving the X-arm from side to side), this becoming worse at mid Z heights than at the ends (top & bottom Z heights).

I've made a spool holder that I put on the floor next to the printer, with the filament fed straight into the extruder rather than via a long PTFE tube. I also mount the extruder on the slot closest to the Z upright rather than the middle slot, so that any pulling from the filament has less effect on the X-arm. Plenty of spool-holder designs on Thingiverse. I quite like this design [www.thingiverse.com]

Dave

I had put the extruder on the first slot as well, I had noticed an improvement doing so at first. I now have the aluminum arm so it's pretty stable, I don't know if it's as important now. But the Z runner has to be tight.
I had printed this spool holder [www.thingiverse.com] but when the spool is too light it doesn't "stick", it gets off of the spool.
I have some 608Z bearings so I'll give the design you gave a try. Thanks.

Quote
mat_fr
I had printed this spool holder [www.thingiverse.com] but when the spool is too light it doesn't "stick", it gets off of the spool.

Maybe weight the spool down by putting a short heavy rod through the hub of the spool, or print some clips that prevent the spool lifting too high..

Dave