Bed construction - a word of warning!

I've just killed my Duet because of a bit of a noddy mistake, but I'm surprised it's not stressed more in the build instructions!

Anyway, I had my aluminium bed plate on the wrong way up, as it's not symmetrical. This meant it was in contact with the solder pad for the bed power supply. I had made about 3 or 4 prints after building with the entire plate effectively live while heating! It was only when doing a large print that The IR probe caught the standard bed clip (I'll be doing the 59p range frame clip mod now!), and pushed it into the M3 bolt at the corner of the plate. Sparks happened, the bed power supply then came through the plate, through the M3 bolt and the clip and onto the IR probe, and back through to the Atmel chip on the Duet which naturally went POOF.

I've contacted reprappro's support about the issue and emailed Sally Bowyer about purchasing replacement parts. Seems like its a really easy mistake to make, so let this be a warning to you!

Hi static65, it may not be entirely your fault, depending on when you built your printer. I made exactly the same mistake when I originally mounted the plate. The original photo showing the bed thermistor being mounted in the middle of the bed showed it facing the wrong way. This means that you have to put the cardboard insulator on the wrong way round too. If you then line up the slot at the edge of the aluminium plate with the one in the cardboard (which seems the natural thing to do), you end up with the aluminium the wrong way round. I reported this at [forums.reprap.org].

Ian said he was going to correct the photo a couple of weeks ago AFAIR.

---

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's a newer printer (I'm assuming) as it's got two holes in the cardboard for the thermistor wires, and I built it on Monday evening. I'm happy to admit fault just thought this is something that should be stressed in the instructions as it's pretty easy to make your plate live and not even realise it!

I've added a warning on the instruction page about this, but haven't been around recently to take the required pictures. We only really worked out that this could be a problem today! It seemed to be a series of consequences just too far, but apparently not. For the record, I worked out what had happened independently...

I think we'll be changing the aluminium heat spreader so it is symmetrical.

Ian
RepRapPro tech support

Perhaps the bed itself should be made to match or more clearance provided to one of the connector slots. My slot (furthest from the Y axis microswitch) seems close to the IDC (assuming I have everything the correct way round!).

---

Ormerod #007 (shaken but not stirred!)

i think it should have been foreseen way before now at the design stage, its called a hazed analysis, professionals use it to analyse their designs before they reach the public domain!

---

# 500

Quote
rm2014
i think it should have been foreseen way before now at the design stage, its called a hazed analysis, professionals use it to analyse their designs before they reach the public domain!

I agree. Putting the plate on the wrong way round is a fairly obvious hazard. I don't understand why RRP didn't make the heated bed pcb symmetrical, then it wouldn't have mattered. Another hazard is having the bed heater terminals exposed at the top of the pcb. Sooner or later, someone (hopefully not me) who is adjusting the bed levelling nuts is going to drop a 5.5mm spanner across them.

---

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].

Quote
dc42

Quote
rm2014
i think it should have been foreseen way before now at the design stage, its called a hazed analysis, professionals use it to analyse their designs before they reach the public domain!

I agree. Putting the plate on the wrong way round is a fairly obvious hazard. I don't understand why RRP didn't make the heated bed pcb symmetrical, then it wouldn't have mattered. Another hazard is having the bed heater terminals exposed at the top of the pcb. Sooner or later, someone (hopefully not me) who is adjusting the bed levelling nuts is going to drop a 5.5mm spanner across them.

I do - RRP have not done any work on the required directives or it would have been done right. (basically I would do the same being that the only exposed lives are 12v) they should have done EMC and electrical (low voltage not so much) also. This product is not CE marked, and therefore should not be sold in the EU, don't get me wrong I am not complaining, I like the idea of new businesses just going for it, and then later kneeling to the man. I do object to launching a product and selling it through RS when its clearly not ready, I think RS have just come to realise this, again I don't care I think this a stroke go marketing genius on RRP's part good luck. Worst that could happen is a short and a fire, hope they have product liability insurance, or someone will lose their house!

---

# 500

It is generally believed that these type of kits do not fall under CE directives and therefore do not require a mark. The issue is not clear cut, and seems to hinge on the interpretation of "product" meaning "manufactured article", although there is no specific exemption for "kits". The vast majority of products sold are finished articles, so it might be natural to assume that "product" means "finished article". I think it is more likely that where the volume of sales of kits is very low it has not attracted regulation.

For some products that are usually sold as a kit (craft toys, construction, medical), CE directives do apply, and it is clear that the definition of "product" does include kits in some areas. The directives there state that when the kit is used or assembled according to the supplied instructions, the final item or use thereof must conform to the relevant CE directives.

Even if a CE mark is not required, then any product sold in the EU must still confirm to the General Product Safety Directive.

The EU directives are unfortunately a lot of red tape for a small company to go through, but perhaps ultimately if it leads to safer products it is not a bad thing.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

Quote
bobc
It is generally believed that these type of kits do not fall under CE directives and therefore do not require a mark. The issue is not clear cut, and seems to hinge on the interpretation of "product" meaning "manufactured article", although there is no specific exemption for "kits". The vast majority of products sold are finished articles, so it might be natural to assume that "product" means "finished article". I think it is more likely that where the volume of sales of kits is very low it has not attracted regulation.

For some products that are usually sold as a kit (craft toys, construction, medical), CE directives do apply, and it is clear that the definition of "product" does include kits in some areas. The directives there state that when the kit is used or assembled according to the supplied instructions, the final item or use thereof must conform to the relevant CE directives.

Even if a CE mark is not required, then any product sold in the EU must still confirm to the General Product Safety Directive.

The EU directives are unfortunately a lot of red tape for a small company to go through, but perhaps ultimately if it leads to safer products it is not a bad thing.

Yes bobc it is a common misconception in reality no such thing exists called a CE directive. If a product is marked with CE it is an indicator of a product's compliance with EU legislation and enables the free movement of products within the European economic area, nothing more. Free trade of products between EU countries has greatly increased the wealth of its citizens and is seen as a good thing. At the heart of this is product compliance with directives and associated testing, perhaps this is where the confusion originates with CE and directives, but both are distinct terms. I can remember a time when just about every country in Europe had its own set of standards, here in the UK they were called British standards, the problem was that to sell your product abroad it had to be certified to each countries standards, that was very expensive and a barrier to trade. Since the 1980s we have had harmonised European standards thank god.

As far as 3d printers go kit or not the law is clear- it is the duty of the manufacturer under HSE legislation, and consumer protection legislation to ensure products sold to the general public meet the relevant directives, it is his sole responsibility to conform to all the legal requirements. Since the design authority (in this case RRP) and the manufacturer are one and the same then any idea that this product is somehow exempt is false, because it is a product which has been intentionally designed to do a certain function, the fact it arrives at the customer as a kit is superfluous. One complicating issue here is the quality of the build, what if the customer is not very handy and causes an accident by not following the instructions correctly, if it came to a dispute over liability then the manufacturer would have to prove his design to be robust and the builder was at fault, one way of demonstrating this would be through product compliance and quality documentation. 3d printers should conform to at least three possibly four directives, and so puts them in a category which when compliance testing is completed, and a technical file produced then a certificate of conformity can be issued and then if you like you can stick a CE mark on it, but its the 3 steps before this that are more important, as an engineer I look for a certificate of conformity which says on it the directives and testing done on a product or component, after all the CE mark alone could just mean China Export!! If I still am not satisfied I will request a copy of the technical file, which will show me in detail the test results.

I would not look on compliance as red tape but rather as good design practice, but it costs quite a lot. Also you don't have to do it (although you would be foolish not to). Should company officers (directors) decide not to carry out product compliance which is entirely possible then it would be a bit like driving a car without a licence, tax or insurance, you can do it but the longer you do it the more the likelihood is you'll get caught out. Avoiding product compliance is too public so if a 3d printer manufacturer does not do it then they are asking for problems and may be asked to take their products off the market, with that comes negative publicity etc etc. Compliance is a good thing, the process of product compliance often uncovers weakness in a design, which is the whole point of it, but if you don't do any then these weakness can be found out by the poor unsuspecting public often to their cost, its at that point that any money saved by the manufacturer not doing compliance is spent on a legal defence up against the HSE, and nowadays company directors can go to prison, if someone were to die for example, perhaps through electric shock.

I must admit I have not looked closely at RRP's documentation to see if they have done any compliance work on the ormerod, i must take a look at it, but if they have it should have shown up this design fault at the HAZARD analysis stage, and if done by a notified body it should have been picked up by their electronics engineers, mechanical protection of live parts is part of it, it is possible that RRP as the design authority mitigated the hazard as low risk, which it is their prerogative to do.

---

# 500

Lets not get too carried away. We are talking 12volts here. Might be a good idea to put an inline fuse on the input lead though !

regards
Andy

---

Ormerod #318
www.zoomworks.org - Free and Open Source Stuff

My concern is shorting the bed heater terminals together, or one of them to ground. Either is likely to cause serious sparks, maybe enough to burn, and may also kill the Duet board. It looks to me that a plastic part could be printed to cover the bed terminal pads. But I may just cover them with PVC tape.

---

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 put a bit of tape on my contacts, just in case a mishap happened.

The PSU, may or may not have short circuit protection but at the power levels and currents other things, wires, connectors, tracks could get hot and the short circuit itself will not be zero ohms and will get hot. I don't recommend a fuse, it's amazing the volts drop they can introduce at normal currents. I suggest a piece of epoxy board (NO COPPER!) made to fit in the gap with a hole to align with the heater pad mounting hole and allow fixing in position.

As for CE marking etc., my understanding is if you take a piece of kit that is CE marked (and I presume not) and make any modification to it yourself, then you take on the responsibilities of the CE compliance, not the supplier or manufacturer.

If this is the case, hands up who has a none modified Ormerod?

As far as I am concerned I'm always prepared to be wrong regarding CE marking as per the differing advice given by 'experts' in the field on specific questions we raised in a previous role. So any experts out there, don't be too harsh!

---

Ormerod #007 (shaken but not stirred!)

Crikey!
Just looked at my bed - and I have that aluminium plate the wrong way round.
Glad I read this topic before disaster struck me.
I know what my next job is....
Greg

---

Ormerod #17

Quote
GregL
Crikey!
Just looked at my bed - and I have that aluminium plate the wrong way round.
Glad I read this topic before disaster struck me.
I know what my next job is....
Greg

So do i!


will be coming off today

Thanks for posting this... I made the same mistake and have printed almost a kilo of things; so have been rather lucky that it's been without incident.
Thanks again... Sharing mistakes is never a mistake

Thanks for the warning. I had my plate the right side up. But just for caution, I placed some Kapton tape on the exposed part of the pins - this way the +12V is insulated and cannot short or create any unforeseen disaster.

Thanks
Sumant

Is everyone happy with the silk screen/solder resist being the only insulator between the heat spreader and the copper traces on the PCB? No doubt it has enough insulation resistance at only 12VDC (so long as it's in good condition). However, with no back EMF suppression we get 50V transients appearing across the heater. I'm not so sure how this might affect things long-term. Another good reason to suppress the transients methinks.

Edited 1 time(s). Last edit at 02/09/2014 05:06PM by Radian.

---

RS Components Reprap Ormerod No. 481

Big thanks to @static65 for highlighting that the aluminium sheet can indeed be the wrong way up...mine, clearly like many others was and is now corrected - I think one that RRP really need to mod the instructions for...

Quote
Radian
Is everyone happy with the silk screen/solder resist being the only insulator between the heat spreader and the copper traces on the PCB? No doubt it has enough insulation resistance at only 12VDC (so long as it's in good condition). However, with no back EMF suppression we get 50V transients appearing across the heater. I'm not so sure how this might affect things long-term. Another good reason to suppress the transients methinks.

Ian told me that the heater PCB is single-sided. If that is correct, the copper traces are on the upper surface (in contact with the glass) only.

---

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].

Hi dc42 the heat spreader in the original design is between the heater pcb and the glass so it is at risk of shorting the tracks of the heater if the solder resist layer breaks down

Quote
Mickyblueeyes
Hi dc42 the heat spreader in the original design is between the heater pcb and the glass so it is at risk of shorting the tracks of the heater if the solder resist layer breaks down

Yes of course - I must be getting brain fade! This could be another reason to dispense with the heat spreader.

Edited 1 time(s). Last edit at 02/09/2014 07:08PM by dc42.

---

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].

dc42 - your brain has been in overdrive making great improvements to Ormorod so we'll let you off

I haven't followed any other discussions relating to the "heat spreader" but, at first blush, it seems like a small enhancement that may well be optional. I'm still digging around for prior knowledge on the topic - I'd hoped to find FLIR type images of with & without etc. but no luck so far.

Edit: I found a FLIR of the heated bed without spreader In this blog post. Still, going off-topic now so maybe better to start a new thread about alternative bed construction.

Edited 1 time(s). Last edit at 02/10/2014 05:40AM by Radian.

---

RS Components Reprap Ormerod No. 481

Just been quoted £110 for a new Duet and IR sensor... huuuuurrrgh. I was happy to admit fault but this is an expensive little error! Didn't realise the Duet was so costly. Hopefully RRP support are feeling nice...

Quote
static65
Just been quoted £110 for a new Duet and IR sensor... huuuuurrrgh.....

Could be even more huuuuurrrgh !

"Duet £119.50"

[www.think3dprint3d.com]

Erik

Ian from RRP has agreed to cover it under warranty, which is a big relief as I don't know how long that would have left my Ormerod out of action for! Always impressed by RRP's level of service