You are not logged in. Please login or register.
SoliForum - 3D Printing Community → Solidoodle Discussion → Odd circle test results
Please let us know what you find out. If there is a solution for the stock carriage, that would be nice, because even if I print replacement parts for the new carriage there's no doubt I'll have to drill the holes to make them circular so it all works, which is lame.
So the thought here (and It may be wrong) would be to prevent the right y-axis carriage from being able to twist during direction changes. After many conversations with jagowilson he had mentioned that the bolt holding one of the 2 Teflon blocks kept coming loose. So I thought what kind of force could cause that? The answer would of course have to be slight twisting motion.
So, since I don't have an SD4 in front of me I can't measure, design and test the theory. I do know that with linear bearings there is no possible way for the x-axis rods to twist in relation to the y-axis rods. What you need to test here is a 3 point design. One block on top of the rod and two (preferably as far apart as possible) on the bottom to replace the existing single point block. All I have to work with here is some rather fuzzy pictures but it would seem to me that the only measurement needed would be the distance apart of the two x-rods, The height between the bottom of the y-rod and the outer extremes of the Y-rod carriage and the width of the rod (8mm's) and then the position of the bolt hole.
I know about the crap bushings and teflon/nylon blocks on SD carriages thats why Lawsy made the replacements with all linear bearings. also the SD crap design is why they failed the space station entry as it allows extra movement when gravity is removed. BTW that same SD design flaw is still being used in the 3 new someday printers of SD also. so they are not just slow learners they are block heads. likely why NASA let them go
Hahahaha. A Solidoodle in space. It doesn't even work on earth ....
yep, my point ;^) plus they want better tech support and safety built in
Hahahaha. A Solidoodle in space. It doesn't even work on earth ....
BWA-HAHAHAHA
This is a substitute for the better solution of installing a third teflon block, which is not possible with the carriage. Does that make sense?
Actually it does make sense 
I'll post something to show if I've gotten you.
What surprises me though and probably why I didn't think this was the issue, is this also the case with the Solidoodle 2 and 3 carriages? None of the Solidoodles can't print proper circles? And this wasn't the first response we got? The Solidoodle staff who make the test prints before shipping (which they probably did once) and the guy in their support video who shows how to fix "your circle problem" doesn't notice this? And somehow I can't notice this issue on the 3d printed parts of the Solidoodle 4 itself?
This was probably why I didn't think this glaringly obvious thing was what you meant.
I don't know how old SDs work but if this fixes it be prepared for an influx of issues when the new printers come out because Facebook pics clearly show the same flaw. Even if it isn't the cause it's a very very bad design and it's sad to see it in the new models. The workbench is just an SD4 with a new Mobo, 2 extruders and lack of an enclosure. If anyone wants that they should get a flashforge because I'm pretty sure it comes with dual E3Dv6s 
As I said in the comments on the fb post what's the point of 2 junk hotends? Why have 2 hotends if you can't use dissolvable support material. Why would anyone care about multi color on the WORKBENCH. They advertise that like it's cool or new to the industry.
And my printed parts show signs of Y backlash. Not sure about yours. OTOH they have bad moire as well so maybe that's what I'm seeing.
Actually,
go to the back of your printer and look for 2 printed black cylinders holding in a screw near the top. Feel it. There are dents. In mine at least. Not sure what these do but they are far from circular even though they look like they should be.
They hold the lid up. See the dent? And no, that's not the start point. 
I'm not sure I see two dents at 5 and 11 oclock. I do see a ton of other issues, including delamation and blobbing (over extrusion) which might have caused that bulge that maybe is the dent you are talking about. Those parts were probably printed pretty fast or with old filament because they can perform their job either way.
Take it off and look at it. They look exactly like the dents in my circles .... exactly. Same size and deviation and everything. Yes there are other issues but the dents are there on my machine all the same. The dents I see when I look at it in the light are backlash-induced ,not caused by over-extrusion.
i cant put this one back on because i dropped the nut somewhere in the printer.... dammit lol
I don't know about how yours look, they are after all 3d printed and not injection molded so might look different on my machine than yours, but I took both of them off and examined them closely. I fail to see two dents at the same offsets. Like I said, I do see delamation and blobbing on mines and any artifact doesn't matter for these two 3d printed parts so I don't think its a good example.
I'm not saying you are wrong, just that I dont see those on mine. If I'd have my camera with me, I'd make close up shots.
Alright. Well it was worth a shot. I'm just trying to find further explanation. We can't be the only ones experiencing this. It just wouldn't make sense. The deviations are so minor that I'm starting to wonder if most people just don't notice . . .
If you can't print proper circles, you won't be able to print proper anything.
I'm by no means an expert either, but after reading this whole thread, I just have to ask: are you guys sure? I remember someone from Solidoodle telling me to make sure the right side of the x carriage has some "free movement". Maybe there's a good reason for it instead? Like maybe to prevent or lessen any damage caused by the extruder head hitting the bed because of bad leveling or bumping on an existing abs on the print bed? (otherwise it would be harder impact)?
I think these stuff need to be asked to Solidoodle first before jumping to conclusion on how crappy the company is and spending the time to replace stuff.
What I'm saying is try out your theories first.
I understand you all are all very frustrated but doing it this way isn't efficient. Good luck.
Everybody seems to think they know more about the issue than the people experiencing it ...
We have contacted Solidoodle on this issue. They assure us that we are just messing up the belt tension. Or that our rod isn't aligned properly. Or that our VREF is off. Or that we're missing sometning else obvious. But let me tell you. I've spent a lot of time on this issue and I very confident in my printer's calibration. I was able to print the parts for the lawsy carriage tonight while I worked. I may not be able to print circles, but my prints are mostly fine. See attached images. This should prove to you that our problem is not typical! Poor circles usually results in many other defects, but the only problem I really have is sloppy corners and slightly non-uniform circles.
Our approach is not inefficient. If we knew of a better solution we'd obviously be doing it. Nothing is alleviating our issues with Y backlash.
It seems we are now deviating in 2 different directions to solve the problem. Jaques seems confident in his ability to design a part to test the 2 points of contact theory. I am going to upgrade to the lawsy carriage. I just dropped about $70 on all the parts. I am replacing every single rod, even the back Y rod.
If either solution results in better circles, we will know this is a design flaw. The SD4 has been filled with them so far; my first real snag was the spring for the extruder being too short and preventing me from using any other filament but Solidoodle's (Octave seems to be more slippery). There are many others I won't dive into, some of which I can't even remember. I see no reason to forfeit the assumption that this is a design flaw.
My parts for the carriages should show up in a few days. When they do, I'll rip out the old rods and carriages, get the hardware store to cut the new rods to size, and install the new carriages and see where I'm at. I will be completely FLOORED if this doesn't fix it, but I bet you it does.
Edit: picture only shows the Y carriage pieces but I'm also replacing X to be thorough. I could get away with just replacing Y, maybe, but I don't want to destroy the old carriages if I can avoid it. First time printing at 100% infill. 
I'm slowly going broke solving this issue. I've almost put the value of the SD4 into it now. Granted, not all my upgrades are related to this issue, but this should give you some idea of what's going on. Our issue is not trivial or typical.
While I wait for parts my side project is to design a replacement Z end stop block. Prior designs for the SD3 don't work on the SD4 because the holes were moved slightly to the left. I've already got most of it done, just need a couple more measurements. I'm motivated to do this because you other SD4 users can't upgrade to the e3d without hacking your endstop like I did to temporarily allow me to print. My hack sucks. I have to balance the bed every print because my endstop screw jiggles. Lol
Even now I can kind of empathize with Solidoodle. Lets be honest: our artifact seems very similar to the y-backlash artifact from belt/pulley issues, its not weird they tell us to try things which solve that.
And the idea that others owning a SD4 haven't noticed and reported this is also pretty unlikely.
Jaques seems confident in his ability to design a part to test the 2 points of contact theory.
Not really. I mean I can design something to add more contact surface to the rod, but I don't know if it can be used for testing, I don't know if ABS will slide as good and I don't know if it won't bring other similar problems which will prevent us to see it as the solution.
The easiest way to me right now seems to be to drill a small hole and add another nylon bushing on the right side of the right carriage part (when looking at it from the side of the printer).
So, I have no idea how similar the carriage is to my solidoodle 2, but on the 2 it looks entirely possible to print something that would hold a couple of LM8UU linear bearings in front and behind the carriage, and remove the teflon blocks, using the bolt hols for them to screw on the new part with the bearings (just a theory I came up with from looking at the carriage, definitely not verified by building the new part, or more likely two parts :-).
What's unfortunate is no one with an SD4 has chimed in to say "I don't have this problem." You'd think that would've happened by now . . .
While it does seem unlikely that SD4 owners wouldn't notice, or Solidoodle wouldn't notice, keep in mind that at least on my machine, this problem is really only strongly evident with your circle test. Tall cylinders have the TINIEST bulge. You wouldn't even see it/notice if you weren't looking for it. The spool mount print really doesn't make it evident enough.
I like the idea of trying to add a linear bearing there. I could probably try that when mine come. Can't hurt. They'll be here Friday I believe. Even if that doesn't work I'll proceed with the upgrade.
And claghorn, if you haven't modified your carriage, could you post pics of the left and right sides, all possible angles please? I don't know how the SD2 was designed but maybe something else will jump out at me.
Just to clear up any confusion, I made a quick youtube video demonstrating the freedom of movement being discussed..
JohnJack has a good point. We need to always include Solidoodle tech support in our progress if we expect to see a refund on the parts we purchase to fix their problem. I've already taken the initiative and sent a query detailing our theories and potential solutions. If they support us, wonderful things could happen.
Wow you can do that with the two Teflon blocks on? Jesus.
Solidoodle won't help you but you're welcome to try. I've been going back and forth with them for months on the issue and they swear up and down were all too stupid to calibrate our belts correctly.
Wow you can do that with the two Teflon blocks on? Jesus.
Solidoodle won't help you but you're welcome to try. I've been going back and forth with them for months on the issue and they swear up and down were all too stupid to calibrate our belts correctly.
Jago, this is without that block you made. Purely Solidoodle's creation.
Edit: and it was only like 2-3 pounds of lift....
SoliForum - 3D Printing Community → Solidoodle Discussion → Odd circle test results
Powered by PunBB, supported by Informer Technologies, Inc.
