Thanks Printbus, that makes a bit more sense.
Chad, do you have a Rumba? I will be waiting in line for a SSR once you get it working! I think I will explore this, and upgrading to Lead screws before changing out the heat bed.
Printable View
Thanks Printbus, that makes a bit more sense.
Chad, do you have a Rumba? I will be waiting in line for a SSR once you get it working! I think I will explore this, and upgrading to Lead screws before changing out the heat bed.
A few closing comments before I try and stay out of this (really, I promise).
An SSR is an electronic device, not mechanical. I've already mentioned there is a functional difference between an SSR for switching an AC load vs DC load. Note that the outputs on an SSR for DC control have a polarity to them. This needs to be considered when you're wiring them up. I don't know from first hand experience if wiring them backwards will damage them, but they certainly won't work that way. Unlike (most) mechanical relays, the control input to a SSR are also polarized.
Being an electronic device, "on" resistance is going to be more of a concern that it was with the mechanical relay. I've read comments of some SSRs people bought for 3D printer heat bed use where they complain the on resistance was unacceptably high, even though their heater drew less than the SSR rated current. The more on resistance the SSR has, the more power you'll lose as heat in the SSR. Some SSRs state they need to be installed on a heatsink to operate at the rated load current, and that internal power dissipation explains why. If the SSRs you are looking at don't specify the on resistance, consider purchasing one for a higher current than you need. A 40 amp SSR will likely have less on resistance than one rated for 25 amps. Yeah, the heftier ones or ones that specify a lower on resistance will cost more.
In retrospect, beefing the hell out of heat bed wiring in order to reduce warm up time may have aggravated the problem. It could be that Marlin was inadvertently sort of optimized around a slower responding heat bed. All it would have taken is for the person developing the control loop to have an inefficiently wired heater.
For those comparing quality between printing PLA on a heated bed and printing on a cold bed, consider that there may be another factor involved in any difference that you see. I know that on some small prints of mine, heating the bed has seemed to add just enough residual heat that the PLA part never properly cooled until the print completed. This meant that the entire print involved layers extruding onto material that was softer than if I printed on a cold bed. Extruding onto soft material is another easy way to end up z artifacts. For example, I know I've seen this happen on 20mm calibration cubes if the infill is high enough. This may not be as much of an issue for ABS, but I've been surprised at what it can take for PLA to cool. Adding more heat from the print surface just makes this worse.
Thanks printbus for all the info. I'm happy to inform that I finally manage to solve my banding issue. . I bought a 3/16 aluminum 6061 plate(a little overkill) And replaced my glass. Also I drilled holes in each corner to fix it with the y plate using spacers with the pcb in between of curse. Since I have abl running already this doest affect my first layer consistency .
This constricted the pcb so it can only flex down on heat cycles without affecting the z height.
My print quality now is great. So definitely the heated bed was causing my banding .
I figured as much, I may go ahead and order one of the MK3 bonded heat beds and being done with it. So basically to fix the issues with the 12" printer you have to rework the Y bed assembly. The majority of the problems I have had with mine are related to the Y bed. Everything from inconsistent Z axis homing, first layer issues, banding, to Y axis crashing.
I would be hesitant to tell anyone to buy one of the 12" printers without getting the aluminum bed and also doing something with the heat bed itself.
Attachment 7124 here is a picture that better illustrate what i did . the only problem of this method is that you need to have a reliable ABL because im not longer able to level my bed .
Also im planning to put the thermistor inside the aluminum bed similar to the ED3d in some place near the edge of the plate so i can actually read the bed temp not the PCB.
Wow, that is a lot of aluminum. Did you have to adjust your acceleration settings or jerk settings at all? I'm thinking if I went that route I would want to take my aluminum bed to a shop and have them mill some holes in it to reduce the weight.
I wonder if there would be a way to mount the aluminum so you could still level the bed. Would clipping it to the glass just defeat the purpose? I like the idea of a PID loop so i will be watching this for a report. If the SSR are as inefficient as it sounds a separate power supply may be needed. My voltage already drops from 13v to 10.5 or so when the heated kicks in and the LED lights flicker quite badly. How much amp draw would just the board and motors take. 10amps? 15?
Both RAMPS and RUMBA are protected to 5 amps on the 12V input that supplies the logic, motors, and extruder heater. RAMBO is a little different in that it separates the power for the extruder heater and fan outputs to an additional 12V input (also limited to 5A).
Having a power supply that fluctuates with load like that is not good.
Nope, it is a power supply from Amazon. I guessing it is just a very cheap unit. So I could power the Rumba, motors, and extruder from a separate 12V 5amp power supply?
Here is a picture of a minion. Never had a problem maybe it was the way it was built. Also have dual extruder. Dual flex. Only problem I always had was when I buy cheap filament. Makerfarm i3v 12inch 6 mm thick glass.
Here are some more picture of minion large and minion key chain I made. The quality is great. Also picture of nasa wrench and dual extruder from itty bitty flex all red is ABS red. And yellow is PLA.
http://imgur.com/aa5adM5
http://imgur.com/SJySlbo
http://imgur.com/cRdz91E
http://imgur.com/bJD551l
http://imgur.com/vT49S1G
http://imgur.com/g3olCbM
http://imgur.com/6rQncjr
http://imgur.com/vW8EDxY
http://imgur.com/v9WgBcj
Red
http://imgur.com/a/anEOv
http://imgur.com/BegTf7w
http://imgur.com/At2aE2Z
http://imgur.com/zCkvZo8
http://imgur.com/I9TzoXh
http://imgur.com/8UXT4Ia
http://imgur.com/bv7dQ4b
That is a upgrade. Hopefully with everything you are doing it will help with the layer problems. I like to stick with pla if i have a choice. I think is easier to print. I too order and received the aluminium plate from colin and as my bed warps but the glass has helped me alot. That is the reason why i bought it. Other problem I do find is if you look at the big minion eyes you would see like waves I don't mine this but this is something i would like to take care of. I haven't tighten my belts or checked them since January this year. Maybe they are loose a little Maybe that problem lies here. I think Printbus documented this.
Also I like to print this color yellow PLA at 215 bed at 60.
Red ABS 240 bed 110.
More ABS prints
http://imgur.com/BegTf7w
http://imgur.com/At2aE2Z
http://imgur.com/zCkvZo8
http://imgur.com/I9TzoXh
http://imgur.com/8UXT4Ia
http://imgur.com/bv7dQ4b
Well I finally got around to measuring my bed warpage with a dial indicator. I wasn't able to get it all the way to the center of the bed to measure it but I am seeing ~.3 mm change in bed height when the heater kicks on. It isn't a gradual change at all and it cycles up and down as the heater turns on and off.
So if I am printing at .2mm layer height the bed is warping more than a full layer each time the heater kicks on and off. My SSR came in today and I am planning on getting it swapped out tonight or tomorrow so I hope that setting up PID will take care of the problem.
So I have printed some test parts with the new aluminum Y bed and found that with the new Y bed and the heat bed on, it has decreased the Z banding quite a bit.
It isn't perfect just yet as there is still some signs of it, so I printed with the heat bed off and found no signs of Z banding issues. My next upgrade was going to be thicker glass. Thank you to rhonal89 for providing your glass thickness!
Glass thickness was what helped a lot. If the heat bed warped the glass held. Also it still heats up quite nicely for abs. I had the same problem like everybody here. Before the glass the heatbed would warp so bad. That leveling the bed was render useless. It would bow down in the middle really bad with thin glass. Also that minion was a 8 hour print. Printed 3 back to back as gifts. I showed them to some guest I had over they said is Disney must be expensive. I said no I printed it. They give it a value of 30 dollars if it was painted at a Disney store
Personally, I'd go for more than 5 amps. Without measuring, I'd assume the input can draw 5 amps, and then it's usually good to have 20-30% margin. The margin is especially important if there's ANY doubt to how much you can trust the rating for the power supply you have. I'd probably see what I could find in the 7.5 amp range or larger. I'd also keep them set to 12V, not 13V, in order to reduce variables I'm adding into the mix.
Here is probably a good fix am also thinking on doing for better prints. Look for a appropriate 12 inch flat heatbed that has it wire solder out of the heatbed not under like the makerfarm I seen some that do this. And clamp it to the aluminum bed and clamp the glass also like make a sandwich some how and use autobed level to level the bed. And counter sink the screws on the aluminium bed so it won't interfere.
Well I got the Solid state relay hooked up and PID setup for the bed.
So far the results are very good, once up to temp the bed only moves ~0.03 mm vs the .3 mm movement before. It seems like PID is the silver bullet for my Z banding, I think there are multiple ways to fix this as have been mentioned in this thread. One good thing about the PID setup is that the bed temp is rock solid and doesn't move off the setpoint.
Here is the relay that I used.
http://www.amazon.com/Lightobject-ES...id+state+relay
Make sure you get a heatsink for it as well as some thermal paste to put in between the two. Also the relay didn't come with mounting screws so you will need a few screws to mount the relay to the heatsink, mine were 4mm for the heatsink I bought. I can't vouch for the long term reliability of the relay as I have only used it to test and to print the cube for comparison.
Here is a comparison between the prints. From right to left, No Heat, Bang Bang "stock setup", and SSR with PID Heat. To me the result is pretty amazing, even the first layer is much better with no voids in it.
Attachment 7167
These were all printed with the exact same settings .2mm layer height, 250° extruder, 110° bed in ABS there were no other changes to the printer other than installing the solid state relay and setting up PID on the bed. Even though I have all the parts to replace the threaded rods and couplers I am not in any hurry to do it because of how good the PID results were.
Hope this helps others out there as I can verify that this fixed my Z banding issue.
I have battled this since I got this printer and IMO a SSR and an aluminum plate for the Y axis should be included with the kit along with instructions to setup PID. I know that the Makerfarm is a kit printer and you should expect to do some tinkering with it to get it printing good parts but as long as you assemble and tune the printer correctly you should be able to expect to produce acceptable prints using the parts that came with the kit.
Thx for sharing your research and results Chadd.
When you did the PID calibration..., how did you store the calibration numbers? In the firmware? NVRAM? ???
That PID cube looks great. I do the same. Every little help in achieving better print quality. As you say my temp changes often also. Also are you telling me ever since you got the printer with the heatbed on your cube would come out like the middle cube.
Also you should open up a new thread and write up a instruction for us all that are looking on doing this. Because it seem cheap and why not do it if it would improve on print quality. Thanks.
Thank You!!! Hope this helps you greatly. And those upgrade you meaning on doing. Will truly helps when you get there.
Wow I get your frustration is you were getting prints like that. Hope this takes care of all those layer problems. And the sound of it if it was because of the heatbed this should help out alot. And now that you say it. I think is the 6mm thick glass that is helping me with those kind of print. Reason is when I have my heat bed temp at 110 I can see it bow below my glass.
I had this same exact issue with my 12" and found a quick solution was to shim up the middle of the bed with cork to remove any room for warping. Not the best solution since it made leveling much more complicated. So I am very excited to see your results and can't wait to try it myself. Great job Chadd!
Here is the thread on PID setup.
http://3dprintboard.com/showthread.p...7322#post67322
You enter them in your Configuration.h file. You also need to enable PID for the heat bed. I started a separate thread on the PID setup.
http://3dprintboard.com/showthread.p...7322#post67322
So I think I too will be making the ssr switch. While I am at it I also think my power supply may need an upgrade. What are you using?
i did not get the unit called out from open builds as they were out of stock. My supply came from Amazon. As stated before, my voltage has drastic swings when the bed kicks in. I'm thinking instead of two supplies I may just get a 40 amp power supply.
I am using a supply that i got from amazon, this is the unit I got
http://www.amazon.com/gp/product/B00...rch_detailpage
Have you checked your voltage at all?