1 Topic by dkeeling728

  • dkeeling728
  • Member
  • Offline
  • Registered: 2013-05-20
  • Posts: 148

Topic: Hotend upgrade?

How difficult is it to change out the hotend to something completely different from OEM. like the "j-head" everyone talks about, or any number of other alternatives?

the reason i ask, is that the USPS managed to lose my replacement/backup parts shipment i ordered from SD a month ago, consisting of a complete hotend, 2 replacement nozzles and 2 heater core assemblies for rebuilding the current one when something inevitably burns out. so while i try to sort this shipping mess out, i want to get something else on order (from a place that ships in a day as opposed to 3 week lead time) as a quick backup in case i have a burnout. im paranoid like that.

from what i understand, i may have to modify the safety temp number in the firmware since other hotends have a more accurate thermistor location to show the real temp. im pretty techy, and i know a bit of C++, enough to get around the firmware at least, but i have no experience at arduino interfacing etc, and TBH, im somewhat hesitant to learn on my printers mainboard. lol.

is the process pretty straight forward? and can i easily swap to the stock hotend or even to another brand fairly easily? as in just comment out the temp line in the firmware for the hotend im not using, etc

and question 2: what is the community consensus on best aftermarket hotend for the SD3? best durability vs size (reduced Z travel)

2 Reply by Rincewind

  • Rincewind
  • Member
  • Offline
  • Registered: 2012-10-16
  • Posts: 334

Re: Hotend upgrade?

I recently switched to JHead and in the end it boiled down to two steps:
1) Printing a MkIV extruder (you can use the jhead with the jigsaw but it's quite some work to adapt and not worth the effort)
2) Replacing the connectors bot from the solidoodle side and the hotend side with JST. You can also find the same connectors as Solidoodle uses (TE Connectivity MTA100) but they are difficult to use without the proper crimping tool.

With the J-Head that I bought (which was not original BTW, but a good clone) I also had to insert a PTFE tube inside the hotend to adapt it to 1.75mm but I didn't need to change the thermistor in the firmware. In general it is worth trying: if you turn on the printer and the temperature you read is roughly the same as for the bed, it should work.

HTH

3 Reply by dkeeling728

  • dkeeling728
  • Member
  • Offline
  • Registered: 2013-05-20
  • Posts: 148

Re: Hotend upgrade?

that's a good thought i didn't think of, common sense that the bed and hotend should both report the same ambient temp when the printer first turns on.

but what i was concerned about was from reading other threads that said some aftermarket hotends report a truer temperature, so that if you set it to 230, the hot end is actually 230, as opposed to setting the SD hotend to 190 to get a true temp of 230 due to the thermistor being simply taped onto the nozzle instead of imbedded in the heater block. meaning that setting the aftermarket hotend to 190 like you do with the stock unit would give you a true 190 and not properly melt some plastic. and setting to 230 would trigger the overtemp protection in the printers firmware.

but i guess im thinking too much, should just go for it and worry about that issue if it appears. don't have anything to worry about at the moment since my printer is still working great with the stock unit it came with. haha

4 Reply by adrian (edited by adrian 2013-07-05 12:20:05)

  • adrian
  • Member
  • Offline
  • From: Canberra, Australia
  • Registered: 2013-01-15
  • Posts: 1,623

Re: Hotend upgrade?

Its not so much a truer or untrue temp...

All it is that the Solidoodle OEM hotend uses the Makergear 'Big Head' Nozzle, and then attaches the thermistor to the exterior and lower down on the nozzle some distance from the heater cartridge. A "standard" hotend has the thermistor located right next to the melt zone / cartridge or resistor.

If you imagine an aircon system.. a Thermistor inside the room being cooled will read much cooler than one outside and in the hall. The thermistor in the hallway has to wait for the cold air to radiate out into the hall before it gets measured. This also means that when the temperature in the hallway is at say 20°C, its probably already down to 15-16°C inside the room. This also means that if you try to control the temperature inside the room using the thermistor out in the hallway, its going to be a lot 'slower' to react to changes than if you measured the thermistor inside the room - the inside room one would see changes instantly almost whereas the one out in the hallway has to wait a while before it 'sees' the changes and knows whether to shut the aircon on or off... meaning if you want the room to be 16°C.. you have to shut the aircon on/off when the hallways thermistor reads 20°C... if you waited for the hallways to get to 16°C, it'd be maybe 8-9°C inside the room with the aircon....

Solidoodle uses thermistors in the hallway. Everyother hotend pretty much has its thermistor inside the aircon'ed room.


(substitute 'aircon' for Heater Cartridge, 'room' for Melt Zone, inverse  the logic from cool to heat, and voila.. you have how PID in your Hotend works.... )

Also - just to note - whilst the standard firmware thermistor is type 6, which is a modified type 1, will 'work' ok with a type 1.. its not 'safe' to just assume if temps match at the low-end then the thermistor tables are 'right' or its 'safe' to use. Thermistor tables are not linear.. this means two different types of thermistor can very much have a similar resistance at a low tempreature but completely different resistances for the same temps further up the table. A thermistor works by varying its resistance as tempreature rises and falls - the firmware has various 'tables' that are used to convert a given resistance to a *known* tempreature and it is in no fashion linear. a 1k change at the low end of the table might be 5°C, but a 1k change at the middle of the table might equate to 20°C differences...

This is why its vital to know *which* thermistor you have - and in actual precision environments, you generally then self-measure the resistance and produce your own 'known good' table - datasheets are close, but every NTC (negative temperature coefficent) or PTC resistor behaves 'differently'.. and thus you usually go through a lengthy calibration process if precision is required instead of 'rough guesstimates' (which is actually all you need in this particular application - so generic tables are 'good enough') We aren't trying to launch a Lunar Mission here - so rough stuff is ok - but don't just assume somethings good to go because ONE point on a table aligns with where you think it should be smile .

Anyway... info provided just to explain why just assuming that because something matches at 20°C ambient to what you think it should be is often bad because it means nothing as you move through the temperatures and up the non-linear scale....

5 Reply by Rincewind

  • Rincewind
  • Member
  • Offline
  • Registered: 2012-10-16
  • Posts: 334

Re: Hotend upgrade?

adrian wrote:

This is why its vital to know *which* thermistor you have - and in actual precision environments, you generally then self-measure the resistance and produce your own 'known good' table - datasheets are close, but every NTC (negative temperature coefficent) or PTC resistor behaves 'differently'.. and thus you usually go through a lengthy calibration process if precision is required instead of 'rough guesstimates' (which is actually all you need in this particular application - so generic tables are 'good enough') We aren't trying to launch a Lunar Mission here - so rough stuff is ok - but don't just assume somethings good to go because ONE point on a table aligns with where you think it should be smile .

Anyway... info provided just to explain why just assuming that because something matches at 20°C ambient to what you think it should be is often bad because it means nothing as you move through the temperatures and up the non-linear scale....

This is all very true and very reasonable, but the problem is that in order to correctly calibrate the thermistor you would need an independent way to measure the temperature, which is not easy to obtain. And on the other hand, with a new hotend you should always try to find out at which temperature your filament melts, because even if you know the behaviour of your thermistor with the greatest accuracy, it is only correlated with the temperature in the melt chamber (as you also said in the beginning of the post).

So my advice is exactly the opposite as adrian's: try the thermistor as it is, if the temperature agrees with the bed, try heating a bit (and hopefully the temperature will go up, if it starts going down it means that the thermistor has an opposite temperature coefficient and you should immediately stop heating). If it goes up, rise the temperature slowly and try extruding until your filament melts properly. And yes, J-Heads in general need temperature settings ~30° above the temperature you use for the stock hotend.

6 Reply by dkeeling728

  • dkeeling728
  • Member
  • Offline
  • Registered: 2013-05-20
  • Posts: 148

Re: Hotend upgrade?

thank you both. i knew it was more complex, but its perfectly clear. will be more apt to experiment before my stock hotend decides to burn up on me, so looking at some alternative hotends now.

im sure i'll have more questions as the experimentation gets started. are there any resources on the web already that explain the calibration process, i would assume the reprap wiki would be a good place to start looking.

7 Reply by Rincewind

  • Rincewind
  • Member
  • Offline
  • Registered: 2012-10-16
  • Posts: 334

Re: Hotend upgrade?

dkeeling728 wrote:

im sure i'll have more questions as the experimentation gets started. are there any resources on the web already that explain the calibration process, i would assume the reprap wiki would be a good place to start looking.

Please feel free to ask... A good starting point could be here: http://reprap.org/wiki/Triffid_Hunter%2 … tion_Guide (Nozzle temperature)