Relay switch addition to prevent heat bed cutting out

I looked at them 3 times, with glasses and a flashlight... there are no markings AT ALL on mine. They are blank yellow.

 

Ohms law is an ideal law and only holds true in certain conditions

But you are right as heat builds, resistance will decrease until ideal conditions are met at which point the current and resistance will equal the voltage. However a lot of that has to do with the supplied amperage.

We need to know the max power of your kapton heaters, that will be the final say in how much current it can draw, especially as a 1.6ohm resistance on your ohm meter could be as high as +-1ohm depending on your multimeter. So going just off of resistance is a little iffy.

You can know for sure by hooking the kapton heater up to the 30A power supply and measuring it's current draw. That will tell you without a doubt if a 15A relay is adequate or if it draws greater than 10A. Or looking at the specs based on energy density and size to determine it's max power and using that value to determine how much current draw it can pull.

As for the temperature control, as the heater uses PWM control that means it cycles max power over a high frequency to give a average power output. So when using a relay you have to allow for max current draw as that what it is allowing to pass through.

 

Ok here's another thing jerry, even if we assume less than 15A load, that relay will never work for a number of reasons.

Even if it's a SSR and not a mechanical relay which I don't think it is because A) SSR's aren't rated for both DC and AC, one or the other, this one is rated for both B) the cycle rate C) it clicks

I had trouble finding this part # GEO 3F-12V3, but I came across this: http://translate.google.com/translate?hl=en&sl=auto&tl=en&u=http://goods.ruten.com.tw/item/show?21112136926880

This lists the max switch rating at no load 300ops/min and rated load of 20ops/min. Currently the heater is running on 10Hz PWM so that means 600ops/min, well above the rate No Load rating. This will die very quickly under given conditions.

Even if the it can withstand the switching rate it's only good for 100,000,000 mechanical cycles, at 10hz, that's around 11.5days of use, not a very long life. The electrical is even worse at 100,000 cycles, or ~2.8hours

We really need an SSR approach.

 

The problem with using a relay, SSR or otherwise, is that you are bypassing the protection. if your heat bed shorts something is going to go up in smoke instead of being protected by the PTC fuse.

Looking at the ramps 1.4 wiki that Cambo posted, that PTC fuse should he this one. Obviously it is working until everything heats up and that is the downfall of this type of protection device. The hold current must be set with the ambient temperature in mind.

I don't have my printer yet so I don't know how close we are talking but I am assuming it is getting warm down there. The hold current for the PTC fuse drops as temperature increases, as you can see from the datasheet, page 138 (RGEF1100). For instance the rating at 25C is 11A, but if things are heated up to 50C then the hold current is only 8.8A and anything much higher than that will eventually cause it to trip (we are pushing around 10A through it???).

What we really need to know is what the "ambient" temperature by this PTC fuse is when the heat bed is running at 100C (or whatever the max temp is for the bed)? Then the proper PTC fuse can be chosen, most likely the solution is the RGEF1400 which is only $1.22.

 

You can use a Transistor if you know how to do it.
A field effect tranistor (FET) would be best, a big one (about 3$) like this one here:
http://www.produktinfo.conrad.com/d...401-da-01-en-Hexfet_Power_Mosfet_IRLU3717.pdf
has an on resistance of 4mOhm => less than 1W power loss at 15 Amps, so that should work just fine.

a high ohm resistor is needed for the base contact to the ground or whatever (otherwise it might doesn't go out after turned on)

 

not true hockey that's why in my mods i've added a circuit breaker on the load side of the ssr to protect against shorts. (shown in the photo below) see my post here http://forums.robo3dprinter.com/index.php?threads/heatbed-and-bed-leveling.841/page-2 and here
http://forums.robo3dprinter.com/index.php?attachments/ssr-heatbed-wire-diagram-v1-png.850/

the control side of the ssr is still protected by the stock resettable fuse.




another problem area that contributes to the temperature issues, is that the robo control board is mounted upside underneath. so any heat given off by arduino ramps board is not convected naturally out. instead the heat goes back into the board AND IT GETS FAIRLY WARM UNDERNEATH. Better way is to mount it right side up.


I've used this board on 3 different printers with no issues. Many others have used it also without problems and this is a first Ive heard of this problem on the ramps board. which leads me to believe the run of boards they are using are not genuine taurino reprapdiscount boards and using subpar parts or there's something with the design of the robo that's causing it... like the mounting upside down.

 

Ahh, didn't know you added a breaker. I didn't see any talk of that from other people though. Still wouldn't it just be easier/cheaper and still have the same out come to swap out that PTC fuse for one rated slightly higher? Or am I missing something???

 

my setup is different than stock. I did my mods for other reasons not related to this thread. You can see my mods in the forums. I was just giving an example of the way I wired up my ssr, but yes replacing it with a better fuse would be my choice too or even a circuit breaker that's not temperature sensitive.
Even better... get that board the proper cooling it needs, mount it right side up and use a fan to keep the board from heating up too much. All the ramps boards come with a fan for a reason. It is to help keep the board cool..

 

Jerry stock firmware has it as PID, if your advice on the relay is to switch to bang bang then you should add that to your OP or else people are going to be trying this approach with stock settings and killing them quickly.

Also if you're using bang bang then the polyswitch is probably good enough.

That polyswitch looks good though

 

Ok, the heatsink is allowing the printer to work MUCH better than before. However, I still cant get it above 100C without it failing. I suspect it is a cohesion issue between the sink and the polyfuse. I did put some thermal paste on the sink, but the fuse isn't anywhere near an even surface. I think I will either replace the fuse with a heavier version, or swap it out for a 10A conventional fuse (or replace it with a breaker). It wont do much for thermal overload, but at least it will protect against a short.