FSR Musings

[mhackney]

What are folks using the FSR sensors using to attach to the leads of the FSR itself? I've tried 3 different 2 pin connectors (.1") and none hole securely.

[Nylocke]

I just solder them and put tiny heat shrink over them. Worked so far for 6 of them.

[mhackney]

Thanks. I ended up doing the same except I soldered on short leads and then installed a 2 pin connector to these leads. It worked well.

I designed some mounts with the "pad" and holder for the FSR for my mini kossel. It has a Smoothie and i want to try 626Pilots' code. I'm using the FSR Endstop Controller, that really makes it a lot easier and the LED green/red feedback for each FSR is nice. I got it all hooked up to test and it seems to be working well.

You'll notice the pad is trlobal, that makes it a lot easier to get a good sliding fit without slop. The triangular bases can can be moved in/out to adjust the radius for the pads to fit into the holders. This worked really well. I have an Orion heated bed and snowflake insulator that overhangs the kossel base.

[mhackney]

Here they are installed. Once I have this working on the mini kossel, I'll do an update for the Rostock. I'm waiting for a couple of Smoothieboards to arrive and one of them is destined for the Rostock Max.

[img]http://mhackney.zenfolio.com/img/s7/v15 ... 7481-3.jpg[/img]

[img]http://mhackney.zenfolio.com/img/s12/v1 ... 0469-3.jpg[/img]

This is the little FSR Endstop Controller with one of the sensors triggered (the red LED). I made a little mount for it so the LEDs are visible.

[img]http://mhackney.zenfolio.com/img/s12/v1 ... 5946-3.jpg[/img]

[mhackney]

Just ran the first calibration with the FSRs on my mini kossel. Worked perfectly!

[mhackney]

Now that I've installed the FSRs and had a chance to play with Smoothie's delta calibration (G29) and 626Pilot's cool Heuristic Annealing Smoothieware branch I have some interesting results.

Firstly, the mounts I made work great. However, I initially just used the PLA plunger with no rubber pad like almost everyone uses. I had sanded the surface of the plunger so it was flat. Running G29 game food results but in 10 tests there would be variations of about 4 steps between the lowest and highest values. Today, at the hardware store, I found 1/2" OD neoprene rubber washers with a 3/16" ID. I figure the surface area of the rubber should be fine so I picked them up. After installing (cemented to the the plunger) I ran a G29 and was astounded that the 10 test measurements were dead on 770 steps for all 10 for a score of "very good"! I did this 3 times just to be sure. This really displays how important some form of soft pad contacting the sensor is. I suppose it spreads the force more evenly over the sensor so there are no hot spots. So, use those pads - even these 1/2 washers are very effective and easily found at hardware stores.

[mhackney]

Just a quick update, after putting the rubber washer above the sensor I've been getting excellent reproducibility - every calibration run is exactly the same # of steps. Can't complain about that. I'm still struggling to get 626Pilot's HA code to run reliably even with the GLCD disabled. But the FSRs are a "win" in my book. I am now designing the mounts for my Rostock and the new Smoothieboards were shipped today so I should be ready to convert it to Smoothie later this week.

[Nylocke]

I had no idea rubber would help repeatability. I've been trying to do a rigid no squish system, I went as far as to use M3 fender washers on my MAX. They seem repeatable enough (haven't tried it out with smoothie so I don't have actual data) but the ABS triggers I made for my schools Mini Kossel were horrible. I was getting anywhere from 4-8 steps difference. Thanks Michael.

[mhackney]

I ran enough tests with the PLA plungers w/o rubber and they were good but not highly repeatable. Adding the 1/2" od rubber washer was an instant improvement. I was very surprised myself!

I got these at the local hardware store - neoprene 1/2" OD, 3/16" ID washers.

[626Pilot]

I just ordered a bag of 100 0.5" OD neoprene washers for something like $10. How are you planning to affix the Rostock mount to the printer?

[mhackney]

Take a look at the STL file I posted about an hour ago in your HA thread. It will screw to the top melamine plate with 2 screws - 1 in each slot. This allows the FSR position to be tweaked so they do not bind. This worked well on the kossel mount I made and this new mount can be used on kossels, rostocks or any printer really.

Screen Shot 2015-03-17 at 7.24.15 PM.png (150.81 KiB) Viewed 10400 times

[626Pilot]

I might modify the mount to clamp to the towers. I don't know if the screw holes are all the same across revisions. Mine look a little different from others I've seen.

[mhackney]

Not sure what you mean by screw holes? I am going to be drilling new holes to attach these to the top melamine plate with some self-tapping screws.

I'm happy to send source files if you'd like. These are Rhino but I could export in most anything.

[626Pilot]

I don't want to drill through lexan. If you can export as STL, that's all I need. Thanks!

[Nylocke]

Isn't Lexan Polycarb? You mean Plexi?

[626Pilot]

Whatever they made my kit out of in Dec. 2012.

[Polygonhell]

Acrylic
Polycarbonate is a very different beast.

[mhackney]

Ok, I already uploaded the STL in the other (your) thread.

[jdurand]

Lexan is a brand name of polycarbonate. Plexiglas is a brand name of acrylic. Kleenex is a brand of facial tissue. And so on...