@DaBit thank you for an interesting reflection on what I assume has been your journey into low cost CNC and the motion control limitations and issues. There is an ongoing shift away from 'industrial' closed loop AC servo drives for light duty machines because of cost. We see Leadshine respond to that with various hybrids of stepper and servo. These are quite good ideas - combine the low cost of laminated pole stepper motors and the ability to avoid a backlash-inducing gearhead or reduction gearing, with refined drives and 4000 tic/rev position feedback. Add step/direction interfacing which still dominates the low end of CNC and automation. And provide local fault monitoring for overtemperature/rotor blocking/undervoltage. This avoids servo tuning, the cost of high bandwidth encoder counting/decoding at the motion controller (Duet card), resonance-free motion, class-leading cost/axis and (it is claimed by Leadshine) sometimes also faster settling times in point to point motion.

@dc42 If Duet 3 is headed into CNC motion control, the codebase will evolve over time, but D I/O optoisolation (via optional CNC-specific tool boards) and CNC compatible signal interfaces and - in time - axis position registers with 6 MHz input channel bandwidth will have to appear. If the ambition is to go beyond hobby-level machines.

What I'd like to see in the near term is a refined step/dir interface for CNC drives and +/- 10Vdc optoisolated D I/O, including example integrations of a few popular drives from Gecko, Leadshine and the like. Implementing this as a compact CAN connected 2-axis toolboard would be a good cost-containment strategy, not least because of the beneficial effect on wiring and connectorisation costs which are frequently substantial.

I'm a new poster so I hope my raising this topic isn't tedious - please be gentle!
Plasticisation and extrusion rate control in FFF printing are more or less open loop processes. In FDM printing (Stratasys) - I don't know whether that too is open-loop. Looking at the injection moulding and extrusion analogues, the norm is provide both 'nozzle' pressure and melt temperature control loops. It would seem reasonable to believe that these would be Good Things if only they were possible. I also read a posting from someone recently concerning use of miniature high temperature piezo pressure sensors with integrated thermocouple from Kistler for hotend melt monitoring.

My initial thoughts:

1. hotend pressure would be modulated about the setpoint by the advancement of the melting filament by the extruder.
2. melt temperature would be maintained at a fixed setpoint as at present, by hotend heaters, or possibly reduced in direct proportion to extrusion rate to allow for shear heating of the melt as it is ejected through the nozzle
3. the setpoint pressure would be modulated according to instantaneous ejection rate. This would probably need to become a low bandwidth autotuned PID control loop in time, due to nonlinearities between extruder motor speed and hotend pressure.
4. if 1 - 3 were possible, it might mean that manual material-specific retraction tuning, 'pressure compensation', monitoring of nozzle bore wear and extruder slip would become much less critical to quality, and might ultimately be relegated to an 'adaptive control' loop via printing of test features at the start of each print. It might also mean that during hotend warmup the rheological properties of the current filament could be inferred, by looking at the relationship between measured melt temperature and pressure.

My questions:

1. if it were practical to do this, could the required extrusion control system be implemented within the planned suite of Duet tool boards?
2. How difficult would it be to provision a few reasonably high resolution noise shielded analog inputs to Duet for industrial pressure sensing (4-20 ma current loop or +/- 10Vdc) to talk to a Kistler charge amp?
3. In the remote event any of this actually worked sufficiently well to improve print reliability and quality, how difficult would it be to develop low cost charge amplifier and signal conditioning circuitry on a tool board that could displace the costly Kistler charge amp and crash the cost of closed loop extrusion?

@ggalisky Re flexures look here:

image url)

This is a non-integrally-formed flexure. Its a large XY table from years ago in my past when I designed a solder paste inspection machine. The Y axis gantry runs top left to top right. The LHS of it is attached to the X-axis. The RHS of it is simply propped, using a thin black stainless steel plate. The lower end of the plate is borne by a linear rail hidden in the black sheetmetal folded section, which in turn is bolted to machined pads welded to the white frame, which is heavy duty box section mild steel. The point of the flexure is to allow the gantry to be a propped cantilever without the front and rear X axis linear ways fighting with each other due to inherent non-parallelism and coplanarity. Vertical and and X-axis stiffness is provided but all other DOF's are unconstrained.

The king of flexures, including integrally-machined ones, is Alexander Slocum, who teaches at MIT. Heres an extract from him: https://mech.utah.edu/~me7960/lectures/Topic12-Flexures.pdf
Another king among inventive engineers with a taste for flexures is Dan Gelbard. His (video) reflections on how to make and use them is here: https://www.youtube.com/watch?v=PaypcVFPs48

Have fun!

Tony Owens

@dc42 You are right David. I was thinking about analog industrial DAQ signal levels for some reason, not digital signals. Automation digital I/O signals were traditionally 24Vdc PNP in Europe, designed to interface to current-sinking controllers (tool board or Duet), because 24Vdc is the common power supply for most low voltage control gear.
I know that CMOS has pushed down the required logic levels of I/O signalling in some equipment.
In any case, what would be best would be to:

1. avoid the need for voltage level-shifting by the user of the tool board
2. provide I/O protection (of the controller, not the driver)
3. avoid the need for providing filtration or snubbing capacitors etc
4. to stick with the bootlace ferrules and screw terminals that are still common for field wiring in the industrial controls industry.
   Obviously what I'm getting at here is the difference in mentality around issues of reliability and availability in the CNC and automation world compared with consumtronics gear. In a situation where a machine is 'printing money' for the user, installation and maintenance has to be slight and slick!

@Danal I recall a discussion involving @deckingman some time ago which touched on the elasticity of the 'shot' of melted (hopefully) material in the hotend, and the effect of this on retraction requirements. There was also theorising about the effect of extrusion rate on the average viscisity of hotend melt, down in the vicinity of the nozzle. The thread was contentious, and I remember thinking that one reason for this was the existence of many variables and influences which were not acknowledged fully. I see this in Moldflow analysis of mould cavities which I occasionally do in my day job, where resin melt heating occurs in regions of high shearing rate (e.g. constrictions), even a long way from the gates where the flow front has cooled.
So, for sure, knowledge of average pressure and melt temperature determines a lot about the rheology of the melted material being extruded from the nozzle, and thus instantaneous extrusion rate. This should help with improving extrusion reliability.

What is the motivation for this? Is it to retain the dynamic performance of the delta effector while greatly shortening the Bowden extruder? If so, surely by approx doubling the moving mass while using the same linear guides and structure and move profiles you risk dropping the lowest resonant frequency of the end effector by about 40%?
Would it not make more sense to stiffen-up the components in the dynamic load loop instead, and just mount a direct-drive extruder to the end-effector? You could use top and bottom parallel motors for sure, or you could:

1. switch to steel-reinforced AT3 or AT5 belts and pulleys and NEMA 23 400 step/rev motors.
2. substitute linear guide profiles that are far stiffer in bending, and do a little triangulation between them if possible
3. replace delta link plastic or magnetic ball joints with either high articulation metal ball joints or universal joints
4. triangulate the main structure to react tower bending and shear loads down into the base.

It seems to me that this sort of issue is a fundamental weakness of Delta printers, because the delta arms are attached to the structure near the top, not near ground level. Unless the structure is also redesigned for larger loads, how does separating the payload into two closely-spaced components moving in parallel going to address the underlying problem?

Tony

I think the susceptibility to jamming of parallel rail systems can be large overcome with minimal loss of stiffness by using parallel flexures in one of the linear rail truck mounts. But only in a system with parallel driving on both linear ways such as CoreXY or twin ballscrew.

I tend to use the simplest and cheapest of IGUS products if at all, due to their attitude to precision. I like them for damping, silence and no need for lube. While other tribopolymer bearing developers like PBC have addressed imprecision of extrusions and heavywall moulded plastic bearing parts in various ways, IGUS resolutely isn't interested in precision plastic bearings combined with reasonable value for money. I had them in to visit with me recently in connection with tower bearings for a heated chamber Delta printer I'm involved with. While they have a range of tribo bearing materials that are rated for service up to 180C, the hard anodised aluminium rails themselves are not suitable for service with these grades, and steel tubes or shafts are needed to exploit them and get a reasonable wear life. I think the material you chose for your sliders is probably a general purpose grade they call J. That is OK to about 80-90 C and works very well with hard anodise.

I'm impressed by the implied quality of printing you can get even on v small TPU parts. Perhaps one reason for that is the relatively stiff structure of your machine, and the structural damping and absence of stick-slip provided by the IGUS bearings?

@gtj0 Did you need to do anything to manage the clearance of those Igus slider parts? Is there enough to be perceptible on the prints? Also, is there any evidence of wear to date, in your use?
Thanks

@ggalisky Re flexures look here:

image url)

This is a non-integrally-formed flexure. Its a large XY table from years ago in my past when I designed a solder paste inspection machine. The Y axis gantry runs top left to top right. The LHS of it is attached to the X-axis. The RHS of it is simply propped, using a thin black stainless steel plate. The lower end of the plate is borne by a linear rail hidden in the black sheetmetal folded section, which in turn is bolted to machined pads welded to the white frame, which is heavy duty box section mild steel. The point of the flexure is to allow the gantry to be a propped cantilever without the front and rear X axis linear ways fighting with each other due to inherent non-parallelism and coplanarity. Vertical and and X-axis stiffness is provided but all other DOF's are unconstrained.

The king of flexures, including integrally-machined ones, is Alexander Slocum, who teaches at MIT. Heres an extract from him: https://mech.utah.edu/~me7960/lectures/Topic12-Flexures.pdf
Another king among inventive engineers with a taste for flexures is Dan Gelbard. His (video) reflections on how to make and use them is here: https://www.youtube.com/watch?v=PaypcVFPs48

Have fun!

Tony Owens

Fantastic project Grayson and given the hands on nature, as distinct from a paper study in sure you are learning a great deal. There's a Masters degree thesis in some of the issues you are touching on as you no doubt are aware.
I'd suggest you use auto coolant in your water cooling system to prevent galvanic corrosion and consider using integral flexures to relieve differential thermal expansion issues without having to add parts.
Be scrupulous about avoiding features also found in Stratasys machines in anything you are considering selling. They are adept at managing their intellectual property and they are not your friends!
We look forward to a refined prototype and to your first PC or Ultem print!

@DaBit Your comments about the limitations of open-loop drives of whatever sort, and the consequences of a mis-sent step (EMI/RFI?) are absolutely true and limit what cheap CNC/automation systems can do. Proper servo's are obviously more ideal in highly dynamic applications. But they have a cost and complexity premium, and for a large class of apps are unnecessary.
I'm very impressed at the range of machine projects you have been involved with. When it comes to flexibility and capability I'm sure LinuxCNC is the best approach short of high end motion controllers (Trio, Galil, Delta Tau) and proprietary CNC packaged systems from Siemens and Fanuc

@dc42 You are right David. I was thinking about analog industrial DAQ signal levels for some reason, not digital signals. Automation digital I/O signals were traditionally 24Vdc PNP in Europe, designed to interface to current-sinking controllers (tool board or Duet), because 24Vdc is the common power supply for most low voltage control gear.
I know that CMOS has pushed down the required logic levels of I/O signalling in some equipment.
In any case, what would be best would be to:

1. avoid the need for voltage level-shifting by the user of the tool board
2. provide I/O protection (of the controller, not the driver)
3. avoid the need for providing filtration or snubbing capacitors etc
4. to stick with the bootlace ferrules and screw terminals that are still common for field wiring in the industrial controls industry.
   Obviously what I'm getting at here is the difference in mentality around issues of reliability and availability in the CNC and automation world compared with consumtronics gear. In a situation where a machine is 'printing money' for the user, installation and maintenance has to be slight and slick!

@DaBit thank you for an interesting reflection on what I assume has been your journey into low cost CNC and the motion control limitations and issues. There is an ongoing shift away from 'industrial' closed loop AC servo drives for light duty machines because of cost. We see Leadshine respond to that with various hybrids of stepper and servo. These are quite good ideas - combine the low cost of laminated pole stepper motors and the ability to avoid a backlash-inducing gearhead or reduction gearing, with refined drives and 4000 tic/rev position feedback. Add step/direction interfacing which still dominates the low end of CNC and automation. And provide local fault monitoring for overtemperature/rotor blocking/undervoltage. This avoids servo tuning, the cost of high bandwidth encoder counting/decoding at the motion controller (Duet card), resonance-free motion, class-leading cost/axis and (it is claimed by Leadshine) sometimes also faster settling times in point to point motion.

@dc42 If Duet 3 is headed into CNC motion control, the codebase will evolve over time, but D I/O optoisolation (via optional CNC-specific tool boards) and CNC compatible signal interfaces and - in time - axis position registers with 6 MHz input channel bandwidth will have to appear. If the ambition is to go beyond hobby-level machines.

What I'd like to see in the near term is a refined step/dir interface for CNC drives and +/- 10Vdc optoisolated D I/O, including example integrations of a few popular drives from Gecko, Leadshine and the like. Implementing this as a compact CAN connected 2-axis toolboard would be a good cost-containment strategy, not least because of the beneficial effect on wiring and connectorisation costs which are frequently substantial.

@Danal I recall a discussion involving @deckingman some time ago which touched on the elasticity of the 'shot' of melted (hopefully) material in the hotend, and the effect of this on retraction requirements. There was also theorising about the effect of extrusion rate on the average viscisity of hotend melt, down in the vicinity of the nozzle. The thread was contentious, and I remember thinking that one reason for this was the existence of many variables and influences which were not acknowledged fully. I see this in Moldflow analysis of mould cavities which I occasionally do in my day job, where resin melt heating occurs in regions of high shearing rate (e.g. constrictions), even a long way from the gates where the flow front has cooled.
So, for sure, knowledge of average pressure and melt temperature determines a lot about the rheology of the melted material being extruded from the nozzle, and thus instantaneous extrusion rate. This should help with improving extrusion reliability.

@bot For order of magnitude estimation, a max push force of 40N (9 lbs) on a 1.75 mm dia filament should theoretically generate up to 97 Bar (1,450 psi) pressure in the hotend (plasticiser barrel). The pressure generated by simple gripwheel extruders is more than many peaople realise!
In reality, some of this filament push force is diverted to tractoring-in the fresh filament to the extruder gripwheels, and notably to overcoming friction between the semi-solid melt and the inner walls of the hotend. But it would appear that the extruder pressure in the hotend can get up to 10-15% of those commonly maintained in the barrel of a plastic injection moulding machine (10k psi at the nozzle of the moulding machine barrel).