Welcome! Log In Create A New Profile

Advanced

Constructive criticism sought

Posted by akhlut 
Constructive criticism sought
February 06, 2014 12:11PM
So I like this x-axis design :[www.thingiverse.com]. The dual-X carriage idea is pretty nifty. Only problem is that I would need to break out some pins and solder up a drive board for the second X-axis. But with access to cheap PCB's from Hongkong I could make my own board, which would be much more satisfying. I could have just bought a RUMBA, but where is the fun in that?

So I whipped this up to give me a second axis.

The Chinese Spring Festival is ending soon, so I have some time to play with the layout.

Any thoughts/opinions/criticism are welcome and encouraged.

Thanks in advance!


- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]
Re: Constructive criticism sought
February 06, 2014 01:53PM
Some nice features there smiling smiley

Are there stepper drivers with the vref on a header pin? There was another guy with a blown board today due to "slipped screwdriver", it would be great to eliminate that weakness from RAMPS.

Isolation on the USB is a neat idea too.


What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license
Re: Constructive criticism sought
February 06, 2014 02:56PM
Thanks!

Yeah. I've followed Roy Cortes and Panucatt for a while now. He makes these drivers that have the fault, decay and vref pins broken out on the board. Might as well use them. smiling smiley

I saw that post. I've done the exact same thing. I finally got wise and ordered a ceramic slotted screwdriver from amazon. For the price of a stepper driver it's a no-brainer. But if you can set the vref with a digipot it's the way to go.

I've read too many people having ground loop issues, so I added the isolated USB. Made up an isolated FTDI for my GRBL controller, figured I should add it to the RAMPS proto board. Works at 9600 for GRBL, and I tested it out at 250000 with a mega the other day (needs a cap on reset to work though) and was able to connect via pronterface, so it should be OK for the board.

The one thing that has me concerned is the ground planes (I am no expert in this or any field). I've broken them out into a logic plane, a motor plane an a FET plane and then tied them together at a single point. Trying to keep noisy things away from quiet things. Not sure if this is better or worse than a contiguous pour.


- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]
Re: Constructive criticism sought
February 08, 2014 05:36PM
Looks good. From just looking at the PCB, some suggestions:

1. Move the stepper boards as close together as possible, and allow for Molex KK connectors for the motor connections (same 0.1" pitch). Even if you don't populate the board with them, it gives you the option. Looks like you're VERY close to fitting them now.
2. Make the stepper motor drive lines as thick as you can make them.
3. Perhaps go for a mini-usb or micro-usb socket? Micro's are all the rage (everyone seems to have one with their phone), but the mini's seem to fall off PCB's less (they tend to have some pins like the shield through the PCB, whereas micros all seem to be SMD).
4. Perhaps a small cap between the USB socket shield and the board ground? (dubious idea, but thought I'd ask).
5. I'd be inclined to rearrange S1, C7 and C8 a bit to give D1 more room (the more it can dissipate, the better), and /or replace D1 with a different component to give better protection (depends on what it is now of course). I'd also make the cathode track from the fuse thicker (from the fuse just to D1's Cathode, not the rest), so that it's definitely going to blow the fuse and not the track.

PS: What fuse holders are those, and what fuses do they take?
Re: Constructive criticism sought
February 08, 2014 07:43PM
Fuse holders look like they are Keystone mini-ATO type [www.digikey.co.uk]

I came to the conclusion that the RAMPS 1N4004 D1 is ineffective, the surge rating is only for a few ms, whereas it takes a lot longer for a PTC or fuse to trip.


What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license
Re: Constructive criticism sought
February 08, 2014 09:21PM
@ Bobc, you're right on the money. Keystone 3568.

And I'm re-visiting the diodes. Eliminating them actually and moving to something else, maybe a MOSFET for reverse polarity protection.

@Cefiar,

thanks for the pointers!

Maybe I'm a curmudgeon, but I like the USB-B connector. I'm kinda leery of those mini and micro connectors. And since I've got the space for it I might as well use it.

And good call on the stepper lines. I'll bump them up to 90 mils.

I tried adding the polarized connectors, but they leave no room as the stepper driver headers are right up against the jumpers. It would be a bear to get them in and out. Ideally I would like to have gone with DIP switches instead of jumpers, but they take up so much room.


- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]
Sorry, only registered users may post in this forum.

Click here to login