Rostock Max V2 ultra level frustrations

[DeltaCon]

Oops, corrected in the posting above...

[captainpotatofudge]

Huh, nope I have not perceived that type of behavior enabling/disabling steppers.

[captainpotatofudge]

If you don't mind, a picture of the cabling before, during, and after the cleanup may also help.

Well here's a couple shots. I tried to get the inside from the sides, but maybe when I get the top plate off I'll get better shots. It's a mess.

[captainpotatofudge]

PCB bed/snowflake/plate lifted. There's the mess.

[gaexcalibur]

well dont feel bad I have had my rostock max 2 for a couple of years now. Finally after trying to sell it on craigs list i threw it into the dumpster. wasted almost 2 grand on this thing i would have been better off with something that worked and was not a delta. bed leveling was my issue as well took 6 hours to level the bed for a single print then it would be off the very next print. I was only on this forum to delete myself when i saw your post and realized i am not the only one who bought one of these lemon printers

[geneb]

It's a poor craftsman that blames to the tool.

g.

[DeltaCon]

Yes, Gene, you are right. But I think it is undeniable that "poor craftsmen" (like myself!) would probably be better off and less frustrated with a cartesian printer. A lot of the succes is dependent on the purpose of your investment. If I needed it for work I would be very frustrated too. In my case I am just interested in the technology and wanting to get a taste of 3d design as an upgrade to my 2d graphical design skills. I must say I learned a lot about technique in the last 6 months, and just now I am beginning to get some confidence in my Rostock again (which I did not build my self, but wish I had!). And I want to become a "good craftsman" badly I hate giving up!

[geneb]

DeltaCon, my reply wasn't directed at you - it was directed at the guy that instead of asking for help, OFFICIAL help, threw (supposedly) his printer in the trash. I'm aware of only one or two cases where the support team at SeeMeCNC have been unable to help a user with a problem. I get really tired of the temper tantrums that get thrown by grown adults when things don't work, right out of the gate. I'm really looking forward to the auto-cal hot-ends becoming available - they won't cure a poorly built machine, but they'll take the pain out of calibrating one that was built with care. (The new v3 design will even help with the sloppily built machines!)

Folks need to understand that while this forum is a pretty big volunteer army of really smart 3D printing nerds, we don't know it all. When an issue comes by that even we can't fix, it's time to email the support gurus at SeeMeCNC. You figure there's probably 4-5 times as many SeeMeCNC printers out there than there are registered users on this forum. We experience a small sampling of the issues that SeeMeCNC is solving every day.

The people most difficult to help are the ones that are sure they know what's going on. The /worst/ tech support calls begin with, "I'm an engineer and..." I suspect the 2nd worst begin with, "I'm a CNC expert and..."

From what I've seen, this is THE best support forum for 3D printers out there and you folks have my utmost respect for the effort you put into helping people, even as I rain snarky commentary down upon some of you.

g.

[Xenocrates]

Gene, I think the "Hello, it's Dr..." calls are worse than the "CNC expert" calls in most cases. Someone with a PHD in engineering would be absolutely terrible to deal with as tech support, if they fit the usual pattern.

Delta, I agree that most people just getting exposed to CNC should be using cartesian systems, as they are simpler to calibrate. However, if you have experience working with almost any CNC, you should be more comfortable working with advanced systems. Admittedly, if all you have is OTJ training, then you may have issues, because you lack the knowledge base to figure out how it works in theory.

[kscott]

So it goes like this: I hit Home All in MatterControl, all carriages move upward at a normal pace nice and smooth, they hit the end stops and immediately let's say the X carriage drops super fast 1/4 of range, the Z drops 1/3 and then ZIPS right back to the top super fast and repeats, while the X slowly moves back up at the same time. Super weird stuff and since it's been totally random so far I haven't been able to capture it on camera. I REALLY wish I could have. And it seems each time it happens if I close MatterControl, the behavior stops immediately.

I had a similar problem when I completed my V2. I traced the problem down to an intermittent connection where the end stop sensor wires connect to the microswitch. I crushed the spade lug connector a little with pliers. Problem hasn't returned... 1 year. A little vibration from the end stop screw hitting the microswitch and a slightly intermittent connection could produce an unreliable home on one or more towers. Hope this helps.

[DeltaCon]

@ kscott: You have your quotes wrong, what looks like my line is actually captainpotatofudge's line. I am not sure he is still reading this forum. But thanks for your suggestion about the endstop wires!

[Eaglezsoar]

DeltaCon, I am glad that you did not give up, and you have contributed a lot to this board.
A personal thank you.

[captainpotatofudge]

I've definitely been so frustrated and angry I've FELT like giving up on it, but in the end I just can't. I may not yet have enough experience or knowledge about delta kinematics required to be a good craftsman in this realm, but overall I consider myself an excellent craftsman in many arenas. I also don't feel like I have the time to invest to learn said delta specific kinematics enough to exactly troubleshoot this beast, thus creating higher levels of frustration for something I expected to function in a specific manner from the start. Which..it sorta did at the start. Then somehow it started degrading ...more..more..

@ kscott: You have your quotes wrong, what looks like my line is actually captainpotatofudge's line. I am not sure he is still reading this forum. But thanks for your suggestion about the endstop wires!

lol I'm still here, but as I said earlier I haven't a load of time to spend on the printer right now so any progress or updates on the rebuild will likely be sporadic. Speaking of which, I have managed to get the towers and entire top plate assembly removed, RAMBO board out and disconnected. The next step which will take more time I don't have yet, is to de-solder or cut all the wiring so I can pull it back through the aluminum extensions to twist them and re-feed them in a more organized manner.

I've also pondered making the stepper wires MUCH shorter. Right now they just seem ridiculously longer than necessary which increases the odds of interference. Anyone have thoughts on this? I'm just trying to reduce as many possible factors as I can since I don't know where my problem stemmed from.

I had a similar problem when I completed my V2. I traced the problem down to an intermittent connection where the end stop sensor wires connect to the microswitch. I crushed the spade lug connector a little with pliers. Problem hasn't returned... 1 year. A little vibration from the end stop screw hitting the microswitch and a slightly intermittent connection could produce an unreliable home on one or more towers. Hope this helps.

That's a good one! not sure that I can think of a way to check for this as they sit, but I'll make sure they're good solid contacts/crushed as I do the rebuild. Thanks for the experience input.

[Xenocrates]

I've definitely been so frustrated and angry I've FELT like giving up on it, but in the end I just can't. I may not yet have enough experience or knowledge about delta kinematics required to be a good craftsman in this realm, but overall I consider myself an excellent craftsman in many arenas. I also don't feel like I have the time to invest to learn said delta specific kinematics enough to exactly troubleshoot this beast, thus creating higher levels of frustration for something I expected to function in a specific manner from the start. Which..it sorta did at the start. Then somehow it started degrading ...more..more..

Don't worry about delta kinematics too much. I studied up on them rather a bit (For example, I have the forward and inverse kinimatic equations somewhere, and tossed them into a spreadsheet for calculations at one point, and I am somewhat familiar with the virtual tower method of kinematic calculation), and most of it is rather useless for this. If the electrical and mechanical build is solid (Which I'm fairly sure you'll manage), then calibration is pretty simple. Adjust H-radius until it moves flat, arm length to get X/Y to line up, and Z height is pretty easy to set with a feeler gauge. It won't be precise to withing thousandths, but you probably don't need that (And if you expect to get that sort of precision out of a hobbiest FDM machine, I have an old bridgeport to sell you)

lol I'm still here, but as I said earlier I haven't a load of time to spend on the printer right now so any progress or updates on the rebuild will likely be sporadic. Speaking of which, I have managed to get the towers and entire top plate assembly removed, RAMBO board out and disconnected. The next step which will take more time I don't have yet, is to de-solder or cut all the wiring so I can pull it back through the aluminum extensions to twist them and re-feed them in a more organized manner.

I know that sort of thing. Going through that a bit right now myself. May I suggest that you pick up some shielded wire for the endstops, as well as the temperature sensor while you're working on it. You'll want to ground the shielding, but it will keep crosstalk from being as much of an issue. Or pair them and twist them before feeding it up the tube. Perhaps some telephony cabling? I would think they make 3 pair cabling that you could feed up the tower as a complete bundle, and have color coded top and bottom for convenience, with a consistent twist for noise rejection.

I've also pondered making the stepper wires MUCH shorter. Right now they just seem ridiculously longer than necessary which increases the odds of interference. Anyone have thoughts on this? I'm just trying to reduce as many possible factors as I can since I don't know where my problem stemmed from.

Hasn't killed me yet. I cut mine down to about the required length. The left stepper in particular had a ton of wire I cut away. big thing would be make sure to either buy new crimp shells (you'll want Molex LP 4 position ones), or save the ones from the cable ends by using an eyeglass screwdriver to release the retention clips on the shells.

Best of luck with things.

[captainpotatofudge]

Man after looking at prices at Digikey for shielded twisted pair...I'll pass! I'll stick to just twisting everything and shortening lengths. That's what I would have done if I had built it anyhow.

big thing would be make sure to either buy new crimp shells (you'll want Molex LP 4 position ones)

Molex LP 4 doesn't look at all right, that's standard ATX computer type plugs. Even with fine tools it doesn't look good for trying to undo the crimp on these. They're simply too small and tight. Looking more like a cut and rejoin with solder job.

[lordbinky]

Bah, there's always braided ground strap you can run cables in and that's inexpensive stuff. Personally I'm thinking of getting some CAT 7 patch cable to chop up and make my wiring nicer. CAT 7 has each twisted pair shielded in foil as well as a braided shield over all the pairs. Even if I go cheap and the wire sizes are a bit small I'll just use multiple pairs together

[captainpotatofudge]

Well I tied into the PSU. Gutted all unnecessary wiring. I left one pigtail lead for 5v in case I should ever need it..no clue right now why I would, but hey...forward thinking, right? The 4 black ground wires going to the first pin were twisted together, then the 4 yellow for the second pin. Then the two bundles were twisted together and connected to the plug. Then a pair of yellow/black twisted together for the 3rd and 4th pins then twisted around the existing large bundle. Then the same for the final pair. Then I Ohm checked each ground termination on the plug to the ground lug on the receptacle and to the PSU case and all read ~0.7-0.8 Ohm.

And I fished the extruder cable out, twisted it and reran through the tower. Then the same for the wires for the PEEK and cooling fans.

I also had an incident with one of the stepper motors where I tried to break torque/threadlock(unsure of how much or what type was used) on a set screw and instead broke a hex bit off at just the right level so as to not be able to get a hold of it with ANYTHING. That took a few days to sort out.

In the meantime I got some Kamigami robots for my daughter and I to play with! They're pretty darn cool!! And with a discount code, they're reasonably priced.

As to getting shielded wire...idk yet. CAT6/7 S/STP or STP isn't terribly expensive I guess, but..idk maybe. I haven't checked, but I'd be surprised if the jacketed shielded CAT cable would it up a tower! So you'd end up just running the shielded twisted pairs separate anyhow.

[captainpotatofudge]

As I'm going through this tear down process I'm noticing that the thermistor wiring for the hot end is SIGNIFICANTLY larger than the therm for the heated bed. What's the reasoning for this? Am I missing something here, because it would be far easier to rout the twisted wires up the tube if the therm wires were smaller gauge. Anyone?

[Xenocrates]

the answer has multiple parts. Firstly, because it's a longer run, the larger gauge reduces inaccuracy, since the temperature is determined by what resistance is sensed, and the larger wire has lower resistance. Secondly, the wiring used for bed power is what the Rambo ships with as a wiring harness for thermistors, while the one used for the hotend comes from a 4 conductor cable that gets the outer jacket removed and is then threaded up. Thirdly, because it is assumed that it will be in near constant motion, and it isn't going to be an easy one to replace, nor do you want any faults with it because of the possibility of fire/damage that results from that, an oversized stranded wire is used.

[gchristopher]

Good luck!!

[captainpotatofudge]

the answer has multiple parts. Firstly, because it's a longer run, the larger gauge reduces inaccuracy, since the temperature is determined by what resistance is sensed, and the larger wire has lower resistance. Secondly, the wiring used for bed power is what the Rambo ships with as a wiring harness for thermistors, while the one used for the hotend comes from a 4 conductor cable that gets the outer jacket removed and is then threaded up. Thirdly, because it is assumed that it will be in near constant motion, and it isn't going to be an easy one to replace, nor do you want any faults with it because of the possibility of fire/damage that results from that, an oversized stranded wire is used.

Quite reasonable! Thanks.