add heat sink to stock hotend

[robgs]

Hi,
Just wondering if it is possible or if anyone has tried to mod the stock hotend for the rostock max v2 to make it high temp compatible? Like remove the plastic components and add a heat sink to the lower portion? I have never taken it apart so i dont know what it looks like disassembled....

[Mac The Knife]

Look up my weekend mods thread. I haz pictures.

[robgs]

Look up my weekend mods thread. I haz pictures.

Thats the spirit Mac!.... i dont know why it didnt come up when i searched before. Youve got some really nice pics there.... do you have some pics of the hotend disassembly?
Im trying to decide on getting the e3d v6 or just upgrade what we have as you've done. Did you just machine a threaded barrel to attach the hotend to the cold end or is thats what is under the plastic heat sink? (Scratch that quesion.. just read through your post now... lol)... Did you have to tap out threads in the upper and lower portion to accomodate the threaded barrel? Did you do anything special to the internals? Hard to tell by the way the pics look...

I was thinking of buying a heat sink for a RC nitro engine similar to this pic

te002r.jpeg (15.36 KiB) Viewed 14176 times

the bottom 15 to 20 mm is solid aluminum and would concentrate the cooling just above the heaters. Simple enough to drill out and mount.... would cost me 12 bucks from the local hobby shop...

but im working on adapting the wasp's resurrection code to our printer so just fishing around for the next possible upgrade...

[Mac The Knife]

The SeeMeCNC parts are threaded 3/8-24, so I threaded the stainless steel tube with the same thread. I did have to add an extionsion I machined out of aluminum to the nozzle, without it the ptfe tube was ballooning at the nozzle.

[Mac The Knife]

And the JB weld I use on the threads burns out after while.

[robgs]

Thanks for the info... I took the hotend apart just to see what I was up against... doesn't look too bad. I haven't found a pic of the innards so here is a pic for anyone who's looking for the disassembly of the hotend.

I didn't have to buy the heatsink as there was one kicking around here. It was a little bit damaged but still workable. I "machined" out the slots for the hotend mount spacers. here's the pics...

I've modified the hotend components to fill in the "air space" that is located between the plastic heat resistor and the nozzle. I filled it with the permatex ultra copper. With this I hope to bridge the gap to allow the heat to dissipate to the heat sink and keep the tip of the PTFE tube from melting down. Hopefully it doesn't dissipate too much heat so as to cool the nozzle. Time will tell... have to wait until tomorrow to test it as I don't have any nylon...

If this doesn't work I will have to order the E3D V6, but hey, that's what this is all about, trying different things out, make things better hopefully... or just end up ordering what you want in the end... lol

I don't know if you've tried any of this or heard of anyone else that is modding their stock hotend. Let me know... it would be good to learn from their mistakes...

[robgs]

Update:
Installed all the components. Heres some pics:

I used the ultra copper to create a solid piece of high temp silicone in the airspace between the heat resistor and the nozzle. I formed it so it would contact the inner thread surface of the hotend barrel when i screwed it in. I used the ptfe tube as a gauge to ensure the nozzle was far enough away from the plastic heat resistor. I just screwed everything together and used a sharpie to mark how far out the tube was sticking. I let i dry overnight.

Above is a pic of the hotend assembled before putting the nozzle on
Below is a pic of the assembled hotend with the heatsink.

I ran a calibration of the assembly at 200°C... and found a major fail... the heaters had a hell of a time getting it to 200.... I had to remove the fan... as soon as i did, the temp shot up to 200 as normal and the calibration continued until the first pid interval... but the temp had a hard time dropping from 208 back to 200 so i had to put the fan back on.... In short, I'll need to tune the fan speed for this heat sink. Problem is I've never done that so I will have to find it in the code and make some mods. If anyone has some helpful advice on getting the fan speed to regulate at a higher temp that would be nice... Until then I'll look through the code....

[teoman]

Or...

Add 2 heater cartridges. More power.

[JFettig]

Is the heat sink attached to the peek or the aluminum? It shouldn't touch the aluminum. Adding RTV to the PEEK where it meets the bottom aluminum piece will suck more heat away from the hot end and into the PEEK, you want less heat into the PEEK.

[robgs]

Thanks for the replies.
Teoman... your idea is probably the answer in this situation lol... needs more energy to sustain the temperature with the huge bleed the heat sink creates. I was hopeful but in this case hope aint gonna cut it.

I found the fan control in the code but i didnt want to start hacking up the code and end up with a burned out mosfet or something like that so i lowered the fan max speed in the eprom settings from 255 to 128... ran a calibration and it worked just fine. I have a thermocouple monitoring the internal temp where the nozzle touches the ptfe tube and i got a couple of readings. At 200 thermistor reading the internal thermocouple was reading about 150 so the energy is being taken away as i hoped. When i went up to 220 the thermocouple showed about 170 to 180 but it took a long time to get up to 220... so i dont think this setup will work with these sized heaters.

Jfettig, the heatsink is attached to the aluminum barrel... i was hoping to divert the energy away from the ptfe tube AND the plastic heat resistor... is that called the peek? Didnt know that. But I could try your suggestion to put the heatsink on the peak, but that means everything below that will see the full temperature. I will try it but what i will do is mould a tube made from the ultra copper and use it instead of the ptfe tube. Ptfe tubing is supposedly rated for 260°C but the ultra copper is good for 370°C so may stand up better. Will give it a go maybe this week....

[JFettig]

PEEK is the material used for the insulator. You're trying to cool the thing you're trying to heat.

[robgs]

Jfettig, thanks for the input. I am trying to apply localized cooling to keep as much heat as possible away from the cold end. Its going to have to be more precise i think. Possibly a barrel that fits over the peek and a small surface area contacting the aluminum instead of how ive got all of the cooling on the aluminum barrel. Also teoman is correct... there needs to be more energy. Thats probably why the e3d v6 utilizes a 40W heater cartridge compared to our dual 3W heating resistors. But just experimenting with cheap alternatives and documenting here so others dont have to make the same mistakes or for those that may have wondered like i did, if its possible... lol... chances are i just end up buying the e3d v6... but its fun just going through the motions...

[JFettig]

Look at the heat break on an E3D hot end. Very little area for the heat to travel up, you need something to isolate the hot part from the cold part, not draw heat out of the hot part, you only end up wasting a ton of energy.

The current hot end is around 48 watts, 3 ohms, 12v, v=i*r and w=v*i.

Adding more heat to the hot end also adds more heat to the heat sink and you do not accomplish what you're trying to accomplish.

[teoman]

I did not fullu understand but, rtv silicone is an insulator. You would be better off filling that part with conductive paste and using the rtv as a gasket between the hotend and your cooler adapter. (I suppose you could cut a piece of silicone sheet aswell).

Also, more power is good but try the upgraded heating element without a fan. In an ideal design the temperature should not shoot too far up. I cannot give an exact temperature number but you do not want it to go above 260. You should be able to find a good compremise between burning your house downd and only damaging the peek after several minutes.

[robgs]

Lol teoman... thats the sweetspot i was going for... just to the brink of burning the house down, then back it off a notch or two...lol
Jfettig... youre right... ohms law is pretty complicated stuff...lol... jk.. i completely understand what you are saying... im making an assumption that the 40W cartridge on the e3d v6 is only rated for 40W but can produce more than 40W... i dont know what the resistive load is there either...but i agree... im going to try putting the heat sink on the peek...

[robgs]

Update:
I did as Jfettig suggested and put the heat sink favoring the peek, but I left about 1 to 2 mm still contacting the aluminum barrel just to help wick some of the heat from the threads of the peek. This time I just left the inside of the barrel clean with no ultra copper. I think this test is a success! Here are some pics.

Above this is the Tcouple temp of the inside just above the point where the PTFE tube and Nozzle meet.

Below this is the actual nozzle temp when Thermistor is reading 230C. the Thermocouple temp is reading 240C.

I have yet to ramp it up to 260 but will give that a go next....

[robgs]

Well.... it didn't go a well as expected, but not because the hotend couldn't do it... going to the extremes, as often happens, brings out some flaws in the machinery.
I didn't want to go to 260C until I had good temperature regulation. From the above pics you can see that my thermistor was reading 10C higher than my thermocouple. So I modified the thermistor table (the userdefined number 5 in the configuration.h) and input the values for our ATC Semitec 104GT-2 thermistor. After doing that I ran a calibration and the temp reading matched much more closely with the thermocouple (within 2C). Also, while I was in there, I changed the Maxtemp value to 275C for the test. I ramped up to 230 without a problem but when I tried to go to 250C the thermistor started spiking and acting all erratic. So I tried a different method of heating. Changed the Heat Manager to setting 3, figured out my dead time and ran that way. anything under 235C runs very well... as soon as I get up to the 240C range the thermistor starts crapping out...
Here's a pic of the graph:

I'm thinking I've got a bad thermistor but maybe there's something wrong with the thermistor table values on RepRap Wiki. Can someone confirm that the values there are OK...
Or I'm doing something wrong? If anyone sees a fly in the ointment let me know.... Short of that I'll change the thermistor and give it another go...
BTW... when I was at 245ish C, the inner temperature got up to 208C... so that's very encouraging... once this is solved I think it should work...

[robgs]

Jeeeeez it takes a long time to get those thermistor lookup tables dialed in properly. Anyways, i changed out the thermistor, changed lookup table to table 8, which contains the more accurate data that matches  ATC Semitec 104GT-2 thermistor. The table on reprap wiki isnt as comprehensive as the number 8 table in our code. It still wasnt perfect but i adjusted the numbers at the 200°C to 260°C to match my thermocouple which was measiring on the nozzle. So that is as accurate as i can get it. Then I ran a M202 S250 and changed the PID to those values. All of this really smoothed out the temperature regulation... no more spikes and randomness...

Heres some pics.

Above, temperature steady state at 260ish with some minor fluctuations and a temp difference of about 1°C from TC to thermistor... im not gonna worry about that small of a difference.

below is the internal temperature, again, where the PTFE tube meets the nozzle. 217°C should be safe to print with nylon.

I'll print a couple of parts as best i can... I'll post some pics of the results.

[Mac The Knife]

After you print, pull your nozzle and check the condition of the ptfe tubing.

[robgs]

Update....
the head prints just fine. The PTFE tube is unaffected. But i have nothing to show just yet. I printed a couple of small parts that are only about 2.5 mm thick, just to test it out. But both prints cracked in half when i applied a little bit of force to them. I dont think im printing at the right temperature or something. First print was at 250C and the second was at 240C. The roll is brand new but maybe i still have to dry it in an oven? Anyways pretty brittle... much more so than ABS. Anyone got any suggestions?
When i print something correctly Ill post some pics....

[robgs]

OK, Here's the pics. I did a 25mm calibration cube. I am not sure what it's supposed to look like but my printer printed it at about 24.6mm all around the middle to top section. The bottom section is pretty much bang on at 25mm... I don't know why it got smaller as it built up. The height is about 25.05mm at the edges but the bottom wasn't exactly flat. The corners kinda curled up a bit but it was very adhered to the bed so just heat warpage I guess.
Mac, here's a pic of the PTFE tube. It's still in really good shape but it did leak the nylon internally... too much pressure I guess. Perhaps have to turn down the feed rate for Nylon. Oh yah, I almost forgot, I am running the 0.35mm nozzle for this print as well...

I'll keep testing it at different temperatures but I think this little experiment is a success. The 3W 6.8Ohm resistors are almost too small for this though. It takes about 3 - 4 minutes to heat up. It stays consistent but the PID tune needs to be extreme at the hottest temperatures in order to control. I've had to set it to P-I-D, 18.65-4.69-75.00 with the PID gain set to 255... Those settings don't work for the lower temperatures as it causes hunting... too bad there weren't settable hightemp lowtemp PID tune buttons... Oh well.. not a big deal to switch out the tune based on what I'm printing....

[robgs]

Its good that I tried this out but is it ever difficult to work with nylon. Its so finicky. It has to be just the perfect conditions for it to work properly. In order to properly test it i will have to put it in an oven for a couple of hours, and build some kind of "dry chamber" to keep it in while its printing. I think it was really brittle because of the moisture it had accumulated while sitting on my machine. I did put it into a desiccant filled bag at night but left the end of it threaded into the machine.
For the parts im making right now it would have been nice to print in nylon but at the moment they wouldnt work. For now im going to keep printing ABS until i have the time to piss around with this stuff. So the final working build looks like this:
-1 rc nitro engine heat sink cut to fit Rostock
-kept everything else stock.. no filling the chamber between the PEEK and nozzle with anything..
-used the stock PTFE tube...
-tested it up to 160°C with no problems other than inability to use layer fan. When the layer fan is used, had a hard time keeping temp.
-heaters could be upgraded to help PID functionality and temp stability.