Question about placement of thermistor on bed.

Nophead why is the thermistor here and not in the center hole where it can touch the glass to get the most accurate reading of what the filament is seeing? [github.com] PG 36.

Besides not being in the hole in the middle may average the temps but they will not be accurate for the other side nor is it safe if you turn it up to 110c so you can print on plain glass while the middle could be over 130c. I ran into someone on the IRC that could not get past about 54c when at the center I bet he was close to 100c touching the glass.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Honestly a 2-3mm alu heat spreader between the glass and heat bed should be recommended. My bed used to heat the middle to 80c and the sides were luke warm. I used thermal paste and aluminum heat spreader to even out the temps. The downside is that the bed now takes much longer to heat up, but its closer to being even. With my thermal probe (multimeter wired one) I probed the bed. It was within 5-6c delta from middle to sides. I find it helps with print quality as well.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

Yeah, I have a heat spreader due in 2 days myself but you know I was shown a flir picture of a silicone mat heater (I believe that is what it was) and I have already seen a MK2A under a Flir and on page 36 that is a MK2x which the Flir pic I saw showed the concentrated heat in the middle where the hole is. So, I was/am a bit shocked to see the placement of the thermistor under the heated bed and not even near the concentration point of the heat for a MK2 based board and a couple of inches/cms away from the hole that was created so the thermistor would touch the glass (or spreader).

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

A properly functioning heated bed should be reasonably uniform in temp across its surface, with colder temps tending towards the periphery.

In terms of the temperature difference between the underside of the heated bed and the topside of the glass, there will always be a temperature difference between the two.

Knowing the exact temperature that the filament is seeing isn't really all that important. What matters is calibration and that your heated bed is consistent (in terms of warming the same way each time, not in terms of warming to the same temp across its area). If your thermistor reads 60°C and you're getting good adhesion, walk away smiling and don't worry about whether the filament is seeing 40°C or 80°C

Don't worry too much about using a specific temperature either - each machine is different, and so are the operating conditions. Just experiment a bit. Up here in Canada, the room my printer lives in is a balmy 16°C, so the temperature difference between the top and bottom of my build platform are probably hugely different. In spite of that, I know that when the thermistor reads 55°C, I get great adhesion.

Just seems wrong for nophead to do it that way.

When I tell a printer I want 115c I want 110-120c. So, having a thermistor off center and under the board seems like a stupid thing to do in my opinion. Especially with the pic on page 36 having it even cemented on to the MK2x board.

This came about from someone on irc needing help and I was giving them help, a long with others, and what he was seeing was nothing like any of us were expecting. Lo and behold it was because of the PG 36 setup which is completely different than most other setups AND what the dang hole was created for in the first place.

As far as your 16c my bed's glass was 12c a couple of days ago (thermocouple reading) so I understand but lets be real and when a complete noobie to all of this is having issues and everyone is helping them based on the norm and up pops a design flawed bed like that excuses are not needed what is needed is for the designer to explain what they were thinking when designing it like that. If a designer can't, or will not, back up what they designed it is time to move away from that design. I know I have always followed nophead as he has a great head on his shoulders so I suspect he will eventually chime in but it will take a lot of explaining for that epoxy/cementing of the thermistor so when it goes out you get to buy a new board but that is a side issue the main one is why the placement where it is when a lovely hole is there made for the job?

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

I agree it is confusing, but in the grand scheme of things its not really going to make or break someone's printer? Just kind of odd...

I'm sure whoever stuck the thermistor there did so for a good reason, and I'm sure their printer works fine, but I certainly can't think of any good reason not to put it in the hole that is provided for exactly this reason? Agreed DarkAlchemist, it would definitely be confusing for someone new.

Interested to hear other perspectives on it?

I am not sure whereabouts the offending thermistor is placed, but assume that it is somewhere off center. My own thoughts on the matter are that the right position is around half way between the center and the periphery. My logic is:

1. Assuming the heater has a constant power input across the whole plate then the hottest point will be in the center whether you are considering heat losses by radiation or convection.
2. A single model being built will occupy the center of the build stage and will insulate this area from cooling
3. Cooling of the model may preferentially cool the middle of the bed but as soon as it gains any height the cooling will be more restricted and again the center will be hotter than the rest of the plate.
4. While allowing that too much heat may be problematic, insufficient heat at any point on the hot plate may promote warping and curl.

In the end though, with so many variables it becomes more of an art than a science. Short of an overwhelming theoretical certainty I would go with the most experienced source - i.e. nophead.

Mike

There are 2 different issues here:

The Mk2 PCB heatbed is a cheap but adequate solution for most people, you will get some temperature differential across the bed, but on a good quality PCB this will not be a big issue. However the majority of cheap MK2's on eBay are not great quality (specifically, they don't have enough copper on the traces), take a look on the Wiki page at Joseph Prusa's warning about cheap beds. I've used both a cheap PCB, and a good quality one, the difference was chalk and cheese. Yes you could use a heat spreader if you want, the additional thermal mass will help to even out temperature fluctuations across the bed (at the expense of longer heat-up times), but with a good quality bed that's not necessary.

Where the thermistor is placed is irrelevant, you are measuring the relative temperature, not the exact temperature on the surface of the bed. It's not practical to measure the top surface temp of the glass with the thermistor, we put the thermistor somewhere under the bed (where, doesn't really matter) then calibrate that measured temperature against the actual surface temp of the glass. You can measure your surface temp with an infrared thermometer, or empirically based on your print results, in my case I know that my thermistor always reads 13 C higher than the bed surface (at normal PLA temperatures), so I can set my bed temps accordingly, if I had placed my thermistor in the hole in the PCB (embedded in a suitable thermal compound) then the only difference I would see is a slightly lower offset.

I completely disagree and I base my disagreement on see the Flir pictures of a MK2. I don't care if the MK2 is crap or golden the fact is the hole in the middle is there for a reason so use it AND epoxying the thermistor to the underside is a really bad too.

"Where the thermistor is placed is irrelevant..." I disagree 100%

I am still waiting for the designer to chime in and hear why he did these things.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.