Welcome! Log In Create A New Profile

Advanced

PCB Effector with embedded sensor

Posted by Lykle 
PCB Effector with embedded sensor
August 03, 2017 08:45AM
I like this idea, a lot!

Effector made out of a PCB, but rather than just using the board material, they have added logic to it. And embedded a sensor in it, to do bed calibration, using the nozzle as a sensing tip.

It uses a new heat sink design my E3D and in my opinion it looks pretty darn good.
Have a gander at the photos and specs.
[duet3d.com]

Lykle
Re: PCB Effector with embedded sensor
August 03, 2017 01:41PM
Looks good. I was one of the beta testers for the new hotend mounting system and I'd love to see where they went with it. I proposed a new adaptable system with modular screw in mount options but I don't think they took it up. I may have to take back the idea!
Re: PCB Effector with embedded sensor
August 04, 2017 01:36AM
Those magnetic arms close to electronics look a bit dangerous to me. In my mind they were only fancy bling-bling, but here they are the only option.
AFAIK, the magnetic circuit isn't closed, so any variation in the airgap between steel ball and magnet will radiate through open air.

It has been tested, so there seems to be no harm. It would probably only disturb bed probing?
The original version was with dc42s IR sensor. I'm sure it would have been a more stable version, compared to the tiny, massively amplified piezo signal.
Re: PCB Effector with embedded sensor
August 04, 2017 01:43AM
What tiny massively amplified signal????????????
Look at first graph on first page of this thread: [forums.reprap.org] 8 Volts at 1mm per second is not tiny.

Mike

Edited 1 time(s). Last edit at 08/04/2017 01:46AM by leadinglights.
Re: PCB Effector with embedded sensor
August 04, 2017 02:15AM
It's missing the option to add spring loaded tensioners between the carriages and the effector. There are holes on the effector which could be used, but they're not in the optimal position. The carriage doesn't have any holes which could be used.
Can the arms get to 90 degrees with the carriage design?
I'd rather see the cold zone portion of the hot end above the effector to lessen tilt effects.
Has anyone done a stress analysis of the PCB for flexing?
Re: PCB Effector with embedded sensor
August 04, 2017 02:22AM
I've tried Traxxas joints, Igus joints and magnetic joints, and of those magnetic joints are by far the best. Also it's been established that having strong springs between the arms (needed by some types of joint) distorts the motion, because they cause the arms to bend which varies their effective length.

The effector doesn't use a piezo for the nozzle contact sensor, it uses a strain gauge. We built a prototype using a piezo as well, but the strain gauge version is easier to manufacture.

The magnetic field from the joints doesn't cause any problems, except to any fans that you mount right next to them. The magnetic field from the hot end fan is a greater potential problem for the electronics, but we have taken precautions against that.



Large delta printer [miscsolutions.wordpress.com], Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].
Re: PCB Effector with embedded sensor
August 04, 2017 03:00AM
Quote
etfrench
It's missing the option to add spring loaded tensioners between the carriages and the effector.

I have used these magnetic joints for a while now and have never needed springs to keep them in, during printing. These magnets are in the arms, so the force is always directly towards the ball and the same strength.
The only time the joints come apart is when I am mounting the extruder or a hot end. Putting force on it to tighten the bolts will pop one off, but that is normal.

Lykle
Re: PCB Effector with embedded sensor
August 04, 2017 03:56AM
Quote
dc42
...........
The effector doesn't use a piezo for the nozzle contact sensor, it uses a strain gauge. We built a prototype using a piezo as well, but the strain gauge version is easier to manufacture.
..............

Strain gauges have a lot of advantages but ease of use by the uninitiated is not one of them. For almost any beginning or intermediate RepRap builders your effector would be a PFM box* - buy, use and don't try to understand - same like smartphone.

Mike

*PFM box -Pure F#####g Magic box
Re: PCB Effector with embedded sensor
August 04, 2017 03:39PM
Quote
leadinglights

Strain gauges have a lot of advantages but ease of use by the uninitiated is not one of them. For almost any beginning or intermediate RepRap builders your effector would be a PFM box* - buy, use and don't try to understand - same like smartphone.

Mike

*PFM box -Pure F#####g Magic box

I'm good with that as long as it works well. I intend to use one on my next Delta build if I dare attempt another one.
Re: PCB Effector with embedded sensor
August 04, 2017 04:47PM
I am not against PFM boxes but am saddened that the RepRap community is moving so far in the direction of approved bits, purchased rather than self built with little or no understanding of the way the bits work and with the same sense of accomplishment as a 12 year old gets when he glues together the five pre-decorated injection molded parts of a model jet fighter.

I really should get with the program and buy a nice shiny Stratasys printer.

Mike
Re: PCB Effector with embedded sensor
August 04, 2017 04:52PM
I suspect most beginning reprappers have a few internet search skills: [en.wikipedia.org]
What is it about a strain gauge that you don't understand?
Re: PCB Effector with embedded sensor
August 04, 2017 05:16PM
Quote
etfrench
I suspect most beginning reprappers have a few internet search skills: [en.wikipedia.org]
What is it about a strain gauge that you don't understand?

I think you misunderstand. I am not the one who doesn't understand strain gauges* I am saying that using standard foil strain gauges is a big ask for beginners - other build it yourself methods including piezo disks are not too bad.

Mike

*Possibly used hundreds in industry including this type [www.rstinstruments.com]
Re: PCB Effector with embedded sensor
August 05, 2017 02:19AM
I must apologize for my half a$$ed comment about it.
I mixed my own bad experience with piezos signal/noise ratio and added some assumptions without reading the real product specs.
Re: PCB Effector with embedded sensor
August 05, 2017 05:55AM
Price? carn't find one anywhere
Re: PCB Effector with embedded sensor
August 05, 2017 07:25AM
Are we taking pcb effectors or strain gauges? Pcb effector and carriages will be £45 plus VAT plus shipping according to Dc42. Which makes it pretty good value, if you have a delta with magball arms, and presuming they work as well as the beta testers say they do. I'll buy one for sure. I'm not sure we will sell any Piezo20 modules to delta owners now, but kits and systems for any/all other printers will still be quite viable.


Simon.

[www.precisionpiezo.co.uk] Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters. PT1000 cartridge sensors plug straight into duet boards and others.
Published:Inventions
Re: PCB Effector with embedded sensor
August 05, 2017 08:23AM
Quote
DjDemonD
Are we taking pcb effectors or strain gauges? Pcb effector and carriages will be £45 plus VAT plus shipping according to Dc42. Which makes it pretty good value, if you have a delta with magball arms, and presuming they work as well as the beta testers say they do. I'll buy one for sure. I'm not sure we will sell any Piezo20 modules to delta owners now, but kits and systems for any/all other printers will still be quite viable.

Not every Delta is easy to switch to Magball arms. I'm putting one of your kits on a Rostock Max V3 and will probably try to figure out how to integrate it into a Tevo Little Monster after I get the Rostock online. Have to have one working printer ;-)
Re: PCB Effector with embedded sensor
August 05, 2017 08:40AM
Cool. I'm not really too worried, its an exercise for me in just getting the tech out there and accepted by more people. The PCB effector looks pretty cool, seems to work and is fairly priced.

I'm expecting a chinese piezo sensor module to arrive on the market any time really....


Simon.

[www.precisionpiezo.co.uk] Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters. PT1000 cartridge sensors plug straight into duet boards and others.
Published:Inventions
Re: PCB Effector with embedded sensor
August 05, 2017 11:56AM
Quote
o_lampe
I must apologize for my half a$$ed comment about it.
I mixed my own bad experience with piezos signal/noise ratio and added some assumptions without reading the real product specs.

No problem. Everybody, including me, were startled how big the output was from piezos. Mostly the noise tends to be mechanical, several people are looking at using digital filtering to engineer this out but recent mods in firmware that DjDemonD has cajoled into existence may make this less useful.

On dc42s effector, I initially thought that strain gauges unsuitable as a DIY RepRap method and would leave most with only the option of buying. Having said that, but a bit of exploration reveals that foil strain gauges will bond straight to FR4 PCBs and work really well - coefficient of expansion, modulus of elasticity and adhesion are all big green tick stuff.

Mike
Re: PCB Effector with embedded sensor
August 06, 2017 08:23AM
im changing my mag ball system to a traxxas ball end as the filament as im using 3mm tends to knock the arms off, possible solution would be to go down to using 1,75mm as it isn't as stiff or hold the tube filament pfte tube with one of those draw string,


Check my rubbish blog for my prusa i3

up and running
[3dimetech.blogspot.co.uk]
Re: PCB Effector with embedded sensor
August 06, 2017 11:09AM
Quote
chris33
im changing my mag ball system to a traxxas ball end

What kind of mag balls were you using? The ones where the ball is magnetic or the ones where the magnets are in the arms and the balls are just steel?

Makes a big difference in attraction force.

Lykle
Re: PCB Effector with embedded sensor
August 06, 2017 11:38AM
no it doesn't a magnet is a magnet depends on the strength og the magnet you use, even you use a draw string to hold your effector on


Check my rubbish blog for my prusa i3

up and running
[3dimetech.blogspot.co.uk]
Re: PCB Effector with embedded sensor
August 06, 2017 12:03PM
Yes but it's apples and orange. With the magnets in the arms they are often cylindrical and much more powerful than cup magnets in the effector which are smaller and less powerful. I've had both systems, it's the Haydn magnet in arm system I've kept.


Simon.

[www.precisionpiezo.co.uk] Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters. PT1000 cartridge sensors plug straight into duet boards and others.
Published:Inventions
Re: PCB Effector with embedded sensor
August 06, 2017 12:46PM
Not completely.
If the magnet is the ball, the force depends on the magnetic field orientation. That can vary how much attraction there is with the arm moving in different positions.

Those joints that are the worst, with the cup magnets on the effector.
There the attraction force is always vertical, while the arm force is always at an angle and sometimes a very acute angle. Bound to release at the wrong moment.

I agree DJ, the Haydn magnet method is the best magnetic system in my opinion. It optimises the strength and always attracts in the correct direction.

Lykle
Re: PCB Effector with embedded sensor
August 06, 2017 01:18PM
you can get different strengths cup and also think a mag sytems ok for aplications such as being able to swap tool heads easy even your picture lykle show your holding the tubing and that using a draw string to take the relief off the effector, as i know myself at far distance it will fall of,also i want to go at speed and it may be a personal approach but i would be more mind at ease that nothing will fall off


Check my rubbish blog for my prusa i3

up and running
[3dimetech.blogspot.co.uk]
Re: PCB Effector with embedded sensor
August 07, 2017 03:31AM
Ah, you mean Brian's delta. Yes as he was using a different material sleeve.
My drive cable is free standing.

I agree completely with you that peace of mind is important, especially with long prints.
Using a string to keep the magnets attached is one way to feel safe.

Lykle
Re: PCB Effector with embedded sensor
August 07, 2017 01:11PM
Quote
Lykle
Ah, you mean Brian's delta. Yes as he was using a different material sleeve.
My drive cable is free standing.

I agree completely with you that peace of mind is important, especially with long prints.
Using a string to keep the magnets attached is one way to feel safe.

Lykle

Earlier you said that wasn't necessary.....
Re: PCB Effector with embedded sensor
August 07, 2017 01:33PM
Cylinder magnet better than cup magnet? I disagree with the examples cited above.

A magnet force is defined by its strength, or gauss. IF you compare a cylinder magnet, and cup magnet, of equal gauss, the magnetic pull will be the same, however you may find more strength in the cup version due to ball sitting in contact with and inside the magnetic cup. If there is an air gap, as is the case in the way cylinder magnets are used in these arms, there will actually be less force, although marginal. In case of doubt, i refer to neo magnets.

I never heard of magnetic ball studs and doubt such a thing exists, mag balls yes, mag ball studs no, because of the way magnets are created. Regarding orientation of magnetic field, in the example cited this is simply not the case, as the ball is in the centre of the magnetic field. Axially magnetised not diametrically.


Flex3Drive.com
Re: PCB Effector with embedded sensor
August 07, 2017 01:54PM
My point was that the gauss is frequently greater with cylindrical magnets in the arms, as this configuration practically allows for larger magnets to be used, which are therefore stronger magnets.

I had a k800 system with 3mm thick cup magnets and arms with steel balls on the end. These magnets were not strong enough fo high jerk/accel, and despite buying N50 (one below the strongest neodymium magnet available) cup magnets to fit this hole in the effector, the performance at extreme arm angles was poor and they disconnected. I tried reinforcing the cup magnets by attaching disc magnets below them to create a stack of magnets increasing the gauss, but it did not work, in fact the disc magnets hanging down from the effector stopped my hotend and cooling fans from turning. I ended up printing magnet mounts to re-orientate the magnets to face the arms more, which definitely improved the performance, as a stop gap while I waited for my Haydn arms/balls to arrive from the states. Adding a bunch more printed parts here definitely made a difference to the accuracy of my printer, which was supposed to be all-metal by design.

Switching to Haydn's magnet system with much larger cylindrical magnets in the arms and I can go up to 350mm/s travel speed with high jerks/accel without disconnections, I am more concerned other parts of the printer will fail mechanically than I am the effector will disconnect from the arms in use.

But as I said above its apples and oranges, were not comparing two equal magnets of different configuration, were comparing two commonly available magnetic arm/carriage/effector systems for delta printers, the magnets on the k800 effector and on Haydn's system are far from equal.

I wrote it all up here [forums.reprap.org] if you can wade through it to find the bit about the magnets. K800 works but not a really high performance or larger machine.

Edited 2 time(s). Last edit at 08/07/2017 01:58PM by DjDemonD.


Simon.

[www.precisionpiezo.co.uk] Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters. PT1000 cartridge sensors plug straight into duet boards and others.
Published:Inventions
Re: PCB Effector with embedded sensor
August 07, 2017 02:18PM
When I first tried magnets I made iron cups with a 10mm concave part that the ball fitted into and 10mm diameter by 10mm long cylindrical N52 magnets. Result was pants - at about 150 grams the ball rode up one side of the cup and disconnected completely at about 350 grams. Replacing the iron cups with brass ones the disconnect was clean at a little over 1kg - this seems completely ridiculous until you try to imagine what the magnetic lines of force are doing. I subsequently replaced the 10mm x 10mm magnets so each had a 10mm diametr by 2mm thick disc and a 6mm diameter by 12mm long - both in N52

Mike
Re: PCB Effector with embedded sensor
August 07, 2017 05:16PM
@djdemond - I didnt realise you were talking about a specific machine. Makes more sense now what you were saying.

Agreed a larger magnet like for like material/spec etc will have capacity for greater forces, but there are a number of variables that make this not such a straight forward assumption, so magnets supplied by a particular company may indeed be weaker, due to being a lower grade. I am not disputing the principles, comments in this thread seemed generalised but stated as absolute, which is not entirely accurate.

@leadinglights, im sure im reading your comment incorrectly. Your not saying brass is magnetic are you. Am I correct interpreting that your iron cups were disrupting a magnetic field reducing its strength?


Flex3Drive.com
Sorry, only registered users may post in this forum.

Click here to login