I’m also going to Vote No for the $700 in Duet hardware.
The Gigabot is printing with current hardware.
The key ingredient seems to be the +effort which is actually a priceless valuable commodity.
Thank You Ryan for demonstrating the beast does work.
Thank You Kevin for documenting an alternative path.
A separate vote $100 for additional parts would have my support.
And some of my +effort, perhaps in assembling an enclosure.
Dave V.
We regularly do upgrades and fixes that aren’t 100% needed to tools, the cnc we added the ability for code to start the spindle when we could just it by hand. The ultimaker just got upgraded to make it easier to use. Laser cutter same thing.
I have talked to multiple people and started a few thread on the mailing group over the last few months regarding the gigabit. The idea of flashing the board has came up a few times but noone has done so. All I can speak for is myself, and I havent because i dont think this is going to fix half the problems out current board has.
From marlin documentation and asking around in forums, the setup we have use between the boards and marlin wont handle the heated bed using a thermocouple while we also use a thermistor for the hot end. Getting the heated bed up is completely possible with the Duet. Another benefit of the duet is being able to speak with people from duet. I was able to show them our setup and they confirmed it can be setup that way.
The other thing that came up over the last few months while talking about fixes is that this arduino board setup was due to the original board not being able to handle the motors and frying and the current set up was used because it’s cheap to replace if it happened again, the duets stepper drivers are designed to handle these motors and bigger.
Between the heated bed and the fact the current setupar felt like a quick fix I still feel that the duet is the better option.
The photo shows we can print in the middle of the bed, but the setup is painful to say the least. Ryan has the most experience with the current setup of the gigabot and has been printing for a few months with limited success on large prints. But arent we trying to make it so everyone can use easier also? The current setup isnt that.
As of right now the rules for voting aren’t set in stone. There is a bylaw vote up to fix that, one gives a time frame, one doesnt. Our current practice however hasnt addressed changes in votes beyond cancellations. But I am going to let this goto as is because I feel these are good upgrades and fixes to the gigabot. If it fails the amazing part of our system is that we can put up a new vote (in all reality it could be the same exact budget for the exact same thing or a new budget completely.)
Woot! How much filament do I need to buy for a giant Darth Vader head???
Nancyg
Depends on how much infill. 
Nancy, At this point the bed is to warped to print something the size of a vader helmet. But a helmet that size is between 1.5 and 3 kg of material.
It could do it. Last month when we tried to get it running for the parts I needed to print for work, we got a part 18 inches by six inches to put down a good first layer.
The only reason I stopped it and switched to the CR10 with a broken up model was because of the configuration quirk where it currently prints things mirrored. While I could have mirrored my models, or corrected the issue, it just seemed to fit my schedule better at the time to use the CR10.
Have you seen pieces printed on a severely warped board like the gigabot? They are garbage.
But Kevin, thank you for pointing out another reason my vote should go through. All these issues like the mirrored prints, the print head randomly dropping on a finished print, the system thinking the nozzle is in a different location than it really is. (You can find alot of this crap on the wiki,it was documented a while back.) All of these were an attempts to fix the gigabot cheaply while using the arduino we have currently. We need something better to get it working properly.
We need a flat print surface, we actually have that but cant use it because of current hardware. We need a system that can handle the stepper motors without a trick to the software. The arduino setup we currently have isnt that. We need a board and firmware that can handle the heated unless we want a system that can only do the most basic filaments.
So are we wanting this to be used by a select few who can muster through or do we want this to be a printer anyone can use? I think what Dave 3.0 said earlier in this vote thread kinda sums up everything: “I want to implore the the most fundamental goal with “improving” any of our 3D printers is that any member can print a basic thingy and it come out looking like a thingy instead of a melted pile of goo.” What we have isnt that. What we have now is a half working antiquated printer that maybe 5 to 10 of the members could force to work which until recently everyone just let sit around.
I, too, am registering my vote of “no” for this purchase, for the reasons Kevin, Dave, and Ryan have already said.
I am registering an official vote of “no” as well. There just doesn’t seem to anything like a consensus that it is necessary and will help. Also it feels to me like a sizable investment is being asked for by a small group of members. As I have mentioned before I feel that little effort has been made to get the 3D printers as “walk up and use by ANYONE” as virtually all of the rest of the areas of the Hive are.
In order to get my vote I would like to see:
Well, I am going to vote “Yes”
As someone who has been building and modifying 3D printers well before you could buy them as kits or even already built I agree with Dan in that RAMPS is a poor choice of processors for a machine of this size. I also agree that a small diameter print in the center of that large bed is in no way indication that everything is fine. Show me a print that high at the limits of the build volume and I’ll change my mind.
After switching from RAMPS to Duet Wifi on both of my FDM prnters there is no way I would ever consider using RAMPS again due to its antiquated setup and limited power.
I agree with whoever said that the printer needs to be walk in and print ready for anyone with little experience and I don’t think it is now.
A new controller is only $165 and there have been offers of donations so maybe we look at doing the minimum to get it to the point where anyone can use it.
I’m gonna vote no on this particular budget. I’d like to see what happens when all the offered parts come in and a little time has been spent trying to optimize it and figure out if and what exactly needs purchased to make this work well. I do appreciate that there are members willing to come in and work on it.
Also, a lot of members have mentioned they’d like a machine that they can walk up to. A better use of the bulk of that $700 might be one or two smaller 3D printers that are on factory settings. Just saying.
What about RAMPS and Marlin makes it a poor choice for this system?
‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
I was wondering the same thing… it’s simple cartesian XYZ motion, not something like a delta that requires extra calculations, and it’s relatively slow because it has so much mass. Ramps and Marlin should handle it just fine.
Printer size doesn’t affect a controller’s suitability at all. The intensity of the calculations depends only on the type of motion and the maximum speed.
Maybe I am missing something, but that is my understanding.
RAMPS / Marlin also is compatible with Trinamic TMC2130 stepper drivers which are quiet and can help manage missed steps.
Just saying, the RAMPS Marlin combo is driving beaucoup printers. It’s well supported with software and hardware updates. Stepper Drivers are stacked, so a blown driver is not a CPU board rework.
If we’re going for the familiar, your average person off the street is more likely to have touched a printer with RAMPS Marlin than any other CPU software combo.
It’s also an order of magnitude lower in cost than Duet.
Dave
Unless I’m mistaken, the Gigabot is not a Cartesian printer, it’s a CoreXY which does require more calculations.
I don’t think the vote is going to pass anyway but my feeling is that the printer would be much more usable with a controller board upgrade and I wanted to register my vote that way.
I was going to reply to Jeff’s comment when I saw yours Kevin. Here are a few of the differences. It’s up to the membership to decide if it’s worth the money to upgrade.
Ramps/Marlin Vs Duet/Reprap
8-Bit vs 32-Bit
Smooth/quiet operation is dependent on the motors and the driver chip’s micro-stepping capabilities. on larger machines, motors and steppers have to work harder. Ramps 8-Bit architecture is limited in that the Stepper Drivers are maxed out pushing the steppers to work harder. This equates to slower less detailed prints. Duet has Stepper drivers with up to 256x microstepping and interpolation mode with Software controllable microstepping
Ramps/Marlin was designed for Cartesian printers. The calculations run by a Cartesian printer’s firmware are much simpler than the ones it needs to run for Delta or CoreXY printers. (Pretty sure the Gigabot is a CoreXY machine)
Duet/Reprap is highly modular with web-based configuration and control. One of the biggest advantages in upgrading is the ease in changing settings. You don’t have to recompile and upload. With the Duet, settings are done in Gcode so you can try a change without having to even reboot.
While auto bed leveling is possible with Marlin, it is a huge pain to set up and get working correctly. Much easier with RepRap firmware and considering there are known issues with the bed of the Gigabot being flat, this would be a big benefit.
With the Duet, you can control the printer from a PC, tablet or smartphone and you can upload gcode files to the SD card on the board via the web interface from those devices.
Duet boards are also built to professional standards, unlike most RAMPS boards which are made by the thousands in unregulated shops.
Duet allows for Higher maximum heated bed current than Ramps, a huge benefit for the large bed the Gigabot has. An upgraded power supply would help heat the bed faster.
Jeff worded it better than I ever could. Everything he mentioned about the standards, interface, modular nature, 256x stepper drivers etc, does apply to Cartesian printers as well as core XY. The duet is one of the best boards on the market at the moment the gigabot can have.
Jeff, currently the heated end doesnt work at all due to the fact of the current board/firmware setup we have. Beyond that it doesnt work from that its setup is pretty decent. Its power is feed seperately from the power supply which helps alot.
All -
I’ve seen a couple misconceptions about my position on this topic, and my recent tinkering with the Gigabot. It seems useful to clarify.
In short summary, I’m not suggesting that the machine is fine as is, and requires no further work. I simply remain unconvinced that $500-700 worth of new parts are required, and I don’t see that the suggested part list addresses all of the primary problems. We might purchase an entirely new 300mm or 400mm printer and repair the Gigabot for the same expense.
Current repair plans seem to have been formulated at a distance based on hearsay and old documentation rather than direct physical examination of the machine. To this end, I printed a small test part on Friday. Before suggesting that the existing electronics still work, it seemed prudent to verify. In further testing Sunday night, a 400mm hexagonal base adhered so firmly that it could not be removed without destroying the underlying tape. Other mechanical (extrusion) problems appeared during this run, preventing completion of the print. The bed, however, while slightly cupped is not as catastrophically warped as we have assumed.
I also found a MAX6675 thermocouple amplifier board hanging off the heat bed, as mentioned in old wiki documentation. I believe, but have not yet fully researched, that this is the missing piece required to re-enable the heatbed.
I’ll avoid belaboring this point by point, but suffice to say a detailed review of the printer would be beneficial before building a parts list and committing member funds to a repair effort. And because I’m uncomfortable being the asshole who torpedoes a vote and walks away without improving things, I will continue to engage in configuring the printer. I’ve had multiple offers of assistance, and there certainly seems to be enough fresh interest in the machine to justify the work.
If the vote passes, I’ll be quite content to stop flailing at the keyboard and move on to other projects. We’re a large group, and there’s always plenty to be done 
P.S. - Jeff S.
I chose your message as a reply target simply because it was the most recent strong summary in the thread. Please don’t consider this addressed specifically at you.
As a series of small and immaterial asides on your comments: The Gigabot is a standard Cartesian machine. 9 point bed probing is already working. The heatbed is an AC silicone mat, so power does not pass through the controller. I’d again agree that the Duet is a better board, and in isolation would remain a fine upgrade for the printer. I remain primarily concerned with total expense, and recommend identifying and following the critical path to avoid unexpected additional expense.
No offense taken Ryan, I haven’t given the machine a hard physical look other than eyeballing it while waiting for the laser. I was basing my opinion on what others have said.
I will disagree with you on the point of the machine being a cartesian printer, if the bed moves in the Z-Axis and the print head moves in the X/Y-Axis then it’s a coreXY machine, but again, I’m going off my memory of what I remember seeing and I’m an old guy with too few brain cells!
Jeff
This vote failed. It was generally stated it could be up and running properly in around a month time by Ryan with manpower. (estimated 2 weeks, plus a cushion to be safe.) I am hopeful this will happen, but I would like to keep a thread open with fixes listed as they happen. We dont need to be stepping on each other toes as we put each of our hands on the pot.
My first recommendation is to use the warden budget to get a BLtouch so we can use the glass board that we already have as well as to see if we can map the curvature of the aluminum board as is.