M2 Auto-Leveling Kit (Beta)
Re: M2 Auto-Leveling Kit (Beta)
So here's a question for the sparkies: why not run the sensor at 24v and then use a dc-dc step down regulator on the output to put it down at 5v? I was thinking of one of these - first mentioned in the stepper damper/quiet fan thread:
http://www.ebay.com/itm/10pcs-LM2596S-D ... 1e88ca6aab
If I go that route I'll have some extras; why not use one of them here?
http://www.ebay.com/itm/10pcs-LM2596S-D ... 1e88ca6aab
If I go that route I'll have some extras; why not use one of them here?
Re: M2 Auto-Leveling Kit (Beta)
The step-down is designed to produce a regulated voltage; that's why it has big electrolytic caps on the input and output. It's not designed to handle a fast-switching input, nor is it really specified for running at 0V, which is what you get when the sensor is off. The better solution is just a simple NPN transistor switch, which then acts very much like the physical contact switch that the Z-min input was previously connected to. It uses a simple pull-up resistor to 5V, and when the sensor turns on, the NPN base is driven high and the collector voltage drops to near 0V, and it all happens very fast. Now probably there's some pre-assembled board you can also get from China that has a 3904 NPN and a pull-up resistor on it for $0.50; I just cobbled together the several discrete components, but it depends on your level of comfort using a soldering iron.
Re: M2 Auto-Leveling Kit (Beta)
Thanks for the reply, Tim. I suspected response time was going to be a problem, but you don't know without asking. It's not my area of expertise.
My soldering is nothing to write home about, so I was just thinking of keeping things tidy - not that daughter boards scattered everywhere is ideal, but I was thinking of an expanded electronics case to hold the extras. I suppose that would work just as well to hide my terrible soldering, so there you go.
My soldering is nothing to write home about, so I was just thinking of keeping things tidy - not that daughter boards scattered everywhere is ideal, but I was thinking of an expanded electronics case to hold the extras. I suppose that would work just as well to hide my terrible soldering, so there you go.
Re: M2 Auto-Leveling Kit (Beta)
I think an electronics enclosure with a prototyping area would be a great idea. But for now, here's what I have. This connects to a Smoothieboard but nothing about it is different from what would go on a RAMBo, apart from the RAMBo having all its 24V power connectors in a row, and not having one hanging in mid-air like I have.pandelume wrote:My soldering is nothing to write home about, so I was just thinking of keeping things tidy - not that daughter boards scattered everywhere is ideal, but I was thinking of an expanded electronics case to hold the extras. I suppose that would work just as well to hide my terrible soldering, so there you go.
I had a piece of perfboard lying around, so I cut it down to this tiny size, which is all that's needed to accomodate the NPN transistor and two resistors that bias the NPN base. For the circuit, see viewtopic.php?f=13&t=2610&hilit=smoothieboard&start=10. You can see the black and blue wires coming in from the sensor (I put a bit of shrink tubing around them to keep things tidier), and the red and other black wire go to ZMIN and GND, respectivly, of the Z-min connector on the controller board. The sensor's brown wire ties directly to 24V on the controller board power input.
Not the prettiest soldering job, but everything is connected together with the component leads and the wire ends.
There's that brown wire, going back to the controller board. I just wrapped the thing in black electrical tape and stuffed it back under the hood, and called it done.
I should be very, very careful to point out that the sensor I have is an LJ12A3-4-Z/BY, not a BX. The BX is a bit easier to work with. There is an internal 10K pullup resistor between the output and the positive voltage supply, so you can use this internal component in place of one of the resistors, and scale the other resistor accordingly. Otherwise, you can use the same switching circuit, but it will work non-inverted, and the Z-min should be specified as non-inverting in the firmware.
-
- Posts: 17
- Joined: Fri Aug 07, 2015 10:37 pm
Re: M2 Auto-Leveling Kit (Beta)
Drew up a v3b holder tried to keep it as far away from the heaterblock as I could.
Re: M2 Auto-Leveling Kit (Beta)
I'm don't think it's necessary to be that far away from the heater block, although time will tell whether or not mine is too close. On the positive axis direction from the hot-ends is the way to go, though. The only good place I could find to mount mine (on the dual extruder) is in front of the hot-ends, and now I have the problem that when it goes to Y=0, the sensor is off of the bed where it can't sense anything. I have not yet figured out what to do about that problem. Several possibilities come to mind, none of them is ideal. Maybe a side mount is best, but on the dual extruder, that puts it pretty far away from the primary hot-end.goopyplastic wrote:Drew up a v3b holder tried to keep it as far away from the heaterblock as I could.
-
- Posts: 17
- Joined: Fri Aug 07, 2015 10:37 pm
Re: M2 Auto-Leveling Kit (Beta)
I wasn't real clear, it's actually in the same place x and y that the v4 holder puts the sensor, I just chamfered it heavily at the base and offset some of the material to keep it away.Tim wrote:I'm don't think it's necessary to be that far away from the heater block.
Re: M2 Auto-Leveling Kit (Beta)
Any chance you'd be willing to share your bracket model? I'm also mounting the sensor to a V3b.
Re: M2 Auto-Leveling Kit (Beta)
I was just trying to justify my own placement of the sensor on the dual extruder. I think it won't melt where I have it, but I'll find out. . .goopyplastic wrote:I wasn't real clear, it's actually in the same place x and y that the v4 holder puts the sensor, I just chamfered it heavily at the base and offset some of the material to keep it away.
Re: M2 Auto-Leveling Kit (Beta)
Tim, does Smoothie have a probe "repeatability" test similar to Marlin's M48? I'd be very interested to see what it reports for the repeatability of the sensor, at 24V input and at 5V input. As I said before, the signal fall-time difference between 5V and 24V input is pretty large, but still not significant on the scale of the machine; I need to rig up my own voltage divider to confirm that, from the firmware's point of view, still.
goopy - all design and testing was done on a v4, sorry for not including a warning; then again, I hadn't intended for it to be used on a dual either, so my intentions and original design goals don't seem to be that relevant here X)
goopy - all design and testing was done on a v4, sorry for not including a warning; then again, I hadn't intended for it to be used on a dual either, so my intentions and original design goals don't seem to be that relevant here X)