[Guide] Using Grbl to debug your homing switches

What would your reasoning be to go with an electrolytic?

Which capacitors? Electrolytics do have a positive and negative leg. On an electrolytic cap, there is an arrow pointing toward the cathod, which is the negative leg. On many electrolytics, the body of the capacitor is black and the arrow is white.

I guess the first reason would be that a 100 uF 50 VDC ceramic capacitor is $32 each.

lol oh, 100. I misread that as 1.00, thinking youā€™d just doubled the capacitance of the first attempt. Yeah, 100uF ceramics are hard to come by. Keeping the 0.47uF ceramic for the second test and putting it in parallel with the 100uF would be a good idea, though.

Hopefully, we want go there. Thereā€™s another option if needed.

1 Like

Well yes, but only if you mean ā€œgroundā€ as in ā€œGNDā€ on the Arduino, which is actually DC Common or 0V.

1 Like

Yup, thatā€™ll work just fine. The capacitors go between the red and black wire, each red gets its own going to the GND pin.

1 Like

is this recommended even if my homeing works well?

@eried

No.

It will slow down the response time to the switches, but only slightly.

If you use soft limits then all you do with the switches is home the machine, thereafter they are ignored.

1 Like

@PhilJohnson

You can run the Arduino without the gShield, no problem.

Disconnect the 24 volt power supply from the gShield ( or donā€™t turn it on).

could be a bad Arduino pin.

Yes. Not likely to be the gShield (it just passes the Arduino pin to the connector), but itā€™s possible.

1 Like

@PhilJohnson the Customer Success team at Inventables is empowered to solve all problems for customers. Toss them a note here and they will get you fixed up:

https://inventables.desk.com/customer/portal/emails/new

1 Like

No. You still need them for your system.

Doesnā€™t hurt anything to leave them in place. It does slow down the response to a switch change, but the slow down is very slight.

1 Like

Possibly, but Iā€™d like you to check out the soldering of the capacitors and the soldering of the connector pin to the gShield.

I canā€™t tell from the picture above where you connected the caps.

@PhilJohnson

Do you have a 10K ohm resistor (or something close to that)?

No. Thatā€™s not really necessary. Right now Iā€™m just trying to figure out what the failure mode is.

This test is what I had in mind. I want to see if the 328P internal pull-up is at issue or if there is really a short somewhere pulling the pin low. There is a 5 volt source on the side of the Arduino where the A5 pin is, but in the other group on that side.

But before you do that, since you canā€™t just pull the connector off, disconnect one of the wires to the Z homing switch to see if the switch is stuck.

Pictures of the 5 pin header installation? Do you have something attached to the header now?

Thereā€™s a lot of possibilities that could explain why you may be having much more noise issues than others such as component tolerances, components bad/going bad, where your board is in relation to other electronics and all sorts of other things.

If you wanted to try @BrianSabanā€™s suggestion the parts are very cheap. If this wasnā€™t an immediate need you could order 100 of each of those parts from China for $3 but it would take a few weeks to get to you. Or you can buy one of each for $.10 and pay $5-10 shipping from Digikey. It would be a real pain trying to solder it up and get it sturdy but you could rig it up and see if it works and then worry about getting it soldered in to survive.

1 Like

Itā€™s late here. Iā€™m shutting down. Iā€™ll think about this some more and get back to you.

1 Like

I read this several times and I just had to smile. To understand why, you need to know that I was a Physics major in college.

Modern Physics is heavily into particle theory. You canā€™t actually see the particles being studied, only the evidence that at one time they were there.

So, that is what made this comment bring back some of the old days for me.

Grounding issues are so nebulous itā€™s usually sufficient to just get rid of them rather than look to see what they are.

To answer your question, I have an Oscilloscope, and an X-Carve. They are not anywhere near each other and I have not taken the time to look at the electrical noise. I just killed it and moved on.:smiley:

1 Like