- User Since
- Aug 31 2016, 12:51 PM (154 w, 3 d)
Thu, Aug 15
@kent there is a surface finish callout in the title block
Wed, Aug 14
the thing is if we have PEEK on the nozzle it would need to melt the PEEK to clean it, which would melt the bristles.
I have been having really good success with using a wiper pad i made that is made from high temp plumbers felt that has been treated with ABS in the same manner as the current wiper pads we use
@logan is going to look at changing the XY steps to ~100.5 to get these to print at the proper size. We are going to be slightly shifting the inserts on the existing parts so the parts are still usable for the alphas. The bowing of the LCD bezel is likely because of the thin printed part. If this is still an issue when the HIPS or PVC bezels come in we will likely want to make a ticket for that and adjust the model to alleviate the problem.
@oliver im going to come over and take a look
@oliver we have had multiple machines with a printed lcd bezel and they do not bow like that. It seems something with the assembly is incorrect
@oliver something is wrong here. i got a set of LCD printed parts from you for the machine that went to carnegie mellon and i inserted them and everything went together smoothly.
@DaniAO last time we got some from top plates from quattro that were kind of matte i believe they were sent back to be reworked. Mechanically all the tool head parts will be fine whether they are matte or have a shiny finish. I personally think they should be fine cosmetically also, but i dont think this is up to me. The back plate is not really visible when the tool head is fully assembled so i think these should be good to use
Mon, Aug 12
@paulette i dont think thats enough. if it is stripping the black coating off then it might be fine. The nozzle should look like shiny stainless steel for good conductivity
@DaniAO i know @marcio developed this when working on kangaroopaw firmware and added it into the TAZ Pro. It is definitely a bug that the screen isnt staying on please wait for the entire duration of the auto homing.
I personally dont think this should hold up production, but that's not really my call
@robert yea definitely! I think we should try and scrub the black coating off the conical portion of the nozzle prior to running calibration. It comes off pretty easily with the green scotch brite pad. Steel wool may work as well. If we can get them from E3d pre cleaned that would be ideal, but it might be a little while until we get there
@robert if you are getting out of round circles then the auto cal isnt getting a good measurement which could be related to T9604 .
If the nozzles are good and clean and can make a good electrical connection with the calibration cube you should not need to touch backlash values.
Also as i mentioned there is no clear way to determine if you need to raise or lower the lash value without using a dial indicator.
@zachah that is weird considering the print takes about twice as long
Fri, Aug 9
@zachah this was sliced using the standard PLA profile for both extruders. Feel free to adjust the slice to increase skirt distance. I do not think we should adjust overall print speed as that was the root of the original problem.
The cake is not printing faster than the original calibration gcode. It is much slower
Thu, Aug 8
mark mentioned concern about the bent pins potentially causing the pcb's to crack.
also as i mentioned, there could be boards that are going to bend them worse than the one we looked at
also with the way the syringe collar is held in place it bows the collar mount slightly when installing the collar and could cause cracks if this is made larger.
we would need to test it and determine the proper amount to apply so it could still be removed if this is the fix we go with
its ethyl cyanoacrylate so it is very close to super glue. but it is specifically formulated for tacking electronics components. and yes the other is an accelerator. We wouldnt want to permanently glue the connector down, but just a tiny dab to hold it in place and so it could still be removed without much effort if need be
@ oliver we need the syringe collar mount to be very strong. i had made them slightly larger before and was getting cracking when the syringe is under a high load. If you just turn the screw when installing through this hole the ribs on the head will shave off a tiny bit of material and it will go through easily
the loctite adhesive stuff seems like the best option in my opinion
@logan it definitely bends it. And there could be boards that are worse than this one
@logan we looked at a 1.3 and a 1.4 rambo and on both boards we looked at the pins will be bent when installing this connector
@logan looking at that with mark, the connector is really tight there and bends the pins. This also would be a lot of reworking rambo boards. @mjpelletier recommends hot glue or the combination of loctite 444 and loctite 7452
@kent that is an off the shelf part. If you want a drawing we will need to contact the supplier/manufacturer
i would like to test these out
It says up to, so does that mean it is normally less than that?
Going off 1,000,000 cycles, a print that is the full z build height with a .5 nozzle will likely have layers somewhere between .1mm and .35mm
if .1 then that is 4500 layers, meaning you could print this 222 times before the tool head needs a new pogo pin board
if .35 that is 1285 layers, meanig you could print this 778 times before needing a new pogo pin board
Wed, Aug 7
is it popping out on the frame side of the chain? on the pivot mount?
cant we just continue producing production units with ABS and recommend to customers to print a shroud out of one of these materials if they will be running the machine for extended periods inside an enclosure. and outfit the cluster machines with blower shrouds printed in one of these materials
@kim the parts from JMT were initial parts to build alpha and beta units since JMT is a fast turn shop.
i would like to get a quote from igus on milling this part and installing the sleeve bushing.
they do this for us on the quiver motor mounts
we decided against printing with glue stick in cluster previously as it would require us to wash parts afterwards
glass transition i think will be more along the lines of the sagging shroud as it is when the material goes from rigid to soft and rubbery which would cause the sagging.
the vicant softening is plunging a small 1mm probe 1mm deep with a set force
PC max should be better based on glass transition temp
i agree standard for both was better
if igus is going to supply this part similar to the motor mount on quiver they can definitely install them. I think we should look into this
@kim would you mind getting a quote from igus on PP-MP0281
@oliver sounds good! thanks for checking that out
updated with c5ecd9790da5
@oliver when you installed the m2x8SHCS into the endstop switch, it looked like that screw threaded fine into the hole, but they were never tapped since it wasnt called out on the drawing. Did you have to force the screw?
The original part did not have the lead nut mounting screw holes tapped, it was assembled with some longer screws and M3 nuts. This allowed a bit of adjustment in case of binding. I then changed the part to have m3 tapped holes on the 5 prototypes and never had a binding issue, but I never updated the drawing to reflect this change. This drawing was sent out to JMT so that is why the alpha/beta parts will not have these holes tapped.
So, the alpha/beta units will have thru holes with the lead nut mounted using 2mm longer hardware and M3 nuts (with loctite). This will function just fine but for production units we should have the tapped holes in the carriage.
It is critical that the countersinks on the lower bearing mount side of the motor mount are countersunk to a 4mm diameter. If this is not the case, the M2x6mm FHCS will not sit flush. This causes the motor to not lay flat when mounted and deflects the motor shaft/worm gear into the worm wheel causing binding and the Nema 11 does not spin freely.
Tue, Aug 6
The installation is incorrect on these. the last photo is the closest.
It looks like the loom wasnt shortened and the motor wire routing is not correct. Also the little printed ears towards the top of the motor should not be there
they should look like this:
Mon, Aug 5
bezel may be a better word than frame
Fri, Aug 2
@adam this needs to go through a change order
Thu, Aug 1
is this PTFE coming from the end of a reel? This setup is no different from TAZ6
This should have been corrected with T7785
@kent what assemblies are you requesting drawings for? every sub assembly? complete machines/tool heads?
does it help if you flip that one the back the other way?
Wed, Jul 31
@logan how much would you be adjusting it by?
what about flipping the feed tube spinner around so the feed side is toward the back? not sure if it will help but it might
Thu, Jul 25
off the shelf https://www.mcmaster.com/99607a120
Its currently under a temporary part number TL-HD0572