Fix to prevent heat bed cutting out using relay switch

The stock F2 polyswitch on the RAMPS board is a Bourns MF-R1100: http://www.bourns.com/data/global/pdfs/mfr.pdf
I believe that this polyswitch is undersized for the job given the temperature de-rating chart.

I replaced it with a Littelfuse 16R1400G: http://www.littelfuse.com/data/en/Data_Sheets/Littelfuse_PTC_16R.pdf.
They are available at Digikey and Mouser for about $1.
http://www.digikey.com/product-search/en?vendor=0&keywords=16R1400G.
It's successfully completed two prints back to back now with my house at 20C, so I think I am good to go. I documented it in case someone else wants to try.

The tools needed were a soldering iron, some desoldering wick, a screwdriver, some tweezers/needle-nose pliers, and some clippers. It was not necessary to disconnect any wiring to perform this fix, and the whole job was done with the Robo3D on its side with the USB port opening facing the ceiling.

Step 1) remove the 3 screws attaching the Arduino control stack to the R3D housing. Move the controller board stack out to where you can work on it:


Step 2) remove the Arduino Mega controller from the RAMPS board. Use your thumb and forefinger to pry each corner apart evenly so that the pins on the RAMPS board don't get bent, and so the crimps in the Arduino sockets don't get bent. Put the Arduino Mega board aside:


Step 3) Clip one of the leads on the F2 Polyswitch, this makes it easier to remove the thing:


Step 4) Use a soldering iron and tweezers or small needlenose pliers to remove the old polyswitch and the clipped pin:


Step 5) Use some de-soldering wick to clean the solder pads. Sometimes solder will stick in the pad holes. You can get it out easily by adding a big blob of new solder with the fat part of the iron and wicking the blob away. It will be easiest if you use leaded solder rather than the RoHS stuff. You should see holes like this when finished. Clean the solder-wick mess up with alcohol and a Q-tip if it bothers you (I cleaned mine):


Step 6) Install the new polyswitch. I got it from Mouser.


Step 7) Clip the leads on the polyswitch a little above the board surface, then solder it in. The spec for through hole stuff is .060", but its fine as long as you can solder to it and it isn't going to interfere with the Arduino Mega board.


Step 8) Press the Arduino Mega board back onto the RAMPS board. Again, try to make it go down evenly so pins and socket crimps don't get bent.


Step 9) Re-install the controller board stack onto the Robo3D housing with the three screws.

 

Thanks Nick, about 4 minutes ago I just bent the F2 part to see what was written ( x 16 and GF1100)

 

Hi Guys,

While my previous adventures in cooling the polyfuses has sort of worked - I can print which was the aim. The reliance on cooling a component to keep it just inside it's operating design is bugging me.

After thinking on the problem I believe isolating the actual PWM control from the RAMPS from the actual supply of high A to the heating elements is the best approach.

My plan is to build a power supply board using Darlington power transistors, heatsinks and a fan. But while I think I know the current and voltage requirements from the discussions on this thread and my own measurements - Can anyone actually confirm what the heat bed and extruder use MAX by design ???

I know I'm basically doubling up on the FET already on the RAMPS but by adding a second layer of control all I risk is a few low cost Darlington's rather than the entire RAMPS board and MEGA.

Im going to assume a 1ms PWM min cycle time.

Thanks - ill upload details once I get further into building the board.

Matt

 

The purpose of the fan is to cool the whole board down not just one component. The ramps/arduino combo comes with a fan for this purpose. Unfortunately, robo failed to install this cooling fan or is using it for another purpose. The mosfets, stepper drivers all give off heat when being used. Which contribute to the board heating up. (the board also being mounted upside down doesn't help either)

anything that has a mosfet or driver is cooled in some way, either actively, passively or even both. It is recommended to use the fan when your operating the ramps board, If you want your electronics to last...

 

@Mark
What did you wire the fan to? Is there fan header on the ramps board?

 

Menos here... Well , I guess you didn't look at the competition... Notice that they use a much larger device for the print pad .



On the other hand I bought a device that was supposed to be functional.... I would think that this should be a warranty retrofit situation , shouldn't it? I realize that the geeks in us want to FIX STUFF, but not good business.

 

is the fan small? Does it fit under the robo with out having to print feet?

 

The fan doesn't fix anything. I was dropping from 100C to 30C between the 15 and 18 minute mark. Now, The bed drops from 100C to ~85C around the 20 minute mark (good enough to complete a print, but have it warp) and refuses to come back to 100C for a second print. I'm with menos50, with more than 10 documented cases of this happening, it's time for Robo3D to step up and actually fix the problem. I don't care if that means shipping out new RAMPS boards with a higher rated resettable fuse soldered in.

 

They have a new board planned for future revisions. I think if you're having troubles with the fan they will ship you a relay or a polyfuse if you request it

My understanding is their whole inventory is in china so they can't fix anything stateside

 

Make sure you are pointing the fan onto the yellow squares.

I have not had any issues and I am using the stock Robo 3d printer.




Sent from my iPhone using Tapatalk

 

jdumbaug,

The initial robo used a different heater for the abs units. The problem didn't arrive until they changed the build plate in October. The polyfuse will generate a lot of heat at 8-9A of power, so even just a slight 25-30C ambient temperature may cause the polyfuse to trip.

So at best they would have caught it around October, and I personally didn't try using ABS for a while. These complaints started coming in and almost everyone found success with a fan solution (which is very common for RAMPS boards I should add). They're now shipping all units with this fan and seems as though it usually works fine.

Now they probably already had an inventory of these ramps boards they need to burn, and replacing that component would be challenging on a lot of units. If the fan is a bandaid that fixes it all the better till they can work out a replacement. Time and money are not the friend of a startup.

There's also no need to be waving your job around like that. People come from different backgrounds and achieve successes in different ways. It's a lot easier in a big company to think of a small problem than it is in a small company to think of a lot of big problems. Being in a "major defense company" might not be as impressive to others as you think it is.

 

Mike,

I definitely didn't know all of that, and was pretty rash to jump to conclusions. I definitely sound like toolbag reading what I wrote! I'm deleting my post, as it certainly didn't help anyone out and isn't the place for that. Sorry all.

 

I just installed the new 14A Polyswitch.....getting the same problems.

 

Hmm, I have the 14A polyfuse and new MOSFET but have not installed them yet. I do have a 60mm fan on the RAMPS board though. My bed can maintain 105C consistently as far as I can tell. I have no good ideas for you right now but I will think about it.

Does the bed switch back on after the polyfuse cools? This would tell you that it is tripping and not something else.

 

Yeah the 14A polyfuse worked when my garage was 10C but now that it's 18-20C it trips as well.

The polyfuse will trigger closed until it has a power reset it seems. It's supposed to reactivate but doesn't seem to

 

Jdumbaugh,

I don't think your problem is actually to do with the polyfuses at all. I have installed a relay switch because my bed was cutting out, and now that doesn't happen now. NOW what happens is what yours does, where it dropps ~15-20C after a couple of minutes. I (and Ziggy) believe that this is due to the fan on the extruder turning on and cooling the bed. As you can see in your screenshot, the bed is being given 100% output but still does not have enough power to overcome this cooling. I have previously had some success bending part of a coke can to make a little fan duct, however I have had some troubles with that and have not yet designed a good printed duct.