Controller Box Problem

Discussion about the CNC Shark Pro Plus

Moderators: al wolford, sbk, Bob, Kayvon

Post Reply
peterchua96
Posts: 70
Joined: Fri Feb 11, 2011 8:42 pm

Controller Box Problem

Post by peterchua96 »

Hi Guys,

It's been a while since I log my account. I encountered problem with my Controller Box. After sometime of usage, the machine stopped in the middle of my work. After checking the unit, I found out that there is a burn in one of the IC inside the controller box. Has anyone of you encountered this problem?

If you guys has an alternative way that you can help me or recommend an alternative third party Controller Box?

As to Next Wave Automation customer support, did anyone experience delay in response?

Peter
Attachments
The burnt chip
The burnt chip
This is where the chip is located in the Controller Box
This is where the chip is located in the Controller Box

Rando
Posts: 757
Joined: Tue Jan 06, 2015 3:24 pm
Location: Boise, ID
Contact:

Re: Controller Box Problem

Post by Rando »

No need for the other pix, I got it. That's an older Shark controller board; my HD2 has the same one....

If you're (quite) good with a soldering iron, it's likely the component can be replaced, but overall the time and part cost will probably be as much as a replacement controller board. Hopefully the board is still under warranty.

That part, C15, is a 100 microfarad electrolytic capacitor, part of the on-board power regulating circuit that drives the digital controller logic and main chip. It looks like they changed to a slightly different kind from the one I have. If your board isn't working, and that component shows signs of burning, then it almost guaranteed shorted out. If you can carefully unsolder just that one part, reconnect everything (don't let it short onto the metal box!) and turn it on briefly. If the lights come on, if it powers-up the motors and such, then you got lucky and it's just that one part. I'd still get the new one, but it's possible to get you back and running quickly. Now comes the tricky part...how to find a 100 microfarad electrolytic capacitor. Any chance you have one of the few remaining radio shacks around? That part isn't all that critical as to it being that tiny, so as long as you get the right value (100 microfarads, working voltage likely ~30V) and make sure the plus-side of the part is toward the OUTSIDE EDGE of the board, it will very likely "just plain work". Once carefully soldered in, it's entirely possible it will entirely fix the problem. If you're up for that but can't the part, let me know and I'll locate one online for ya ;-).

But again, if it's under warranty, a new board is really the best option.

Now, as to delays in support, yes, I think we can agree that such delays are occurring with unwelcome frequency. :(. Given you already know what's wrong, maybe a call to their sales line would get you in the queue faster; they like sales :D.

Regards,

Thom
=====================================================
ThomR.com Creative tools and photographic art
A proud member of the Pacific Northwest CNC Club (now on Facebook)

User avatar
Kayvon
Posts: 552
Joined: Tue Oct 21, 2014 11:46 pm

Re: Controller Box Problem

Post by Kayvon »

Rando wrote:That part, C15, is a 100 microfarad electrolytic capacitor, part of the on-board power regulating circuit that drives the digital controller logic and main chip.
You sure it's not a 22uF 35V capacitor? Look at page 3 of Vishay's datasheet: http://www.vishay.com/docs/40002/293d.pdf
Rando wrote:If you're (quite) good with a soldering iron, it's likely the component can be replaced, but overall the time and part cost will probably be as much as a replacement controller board. Hopefully the board is still under warranty.
If you just want one, it's a $1 part: https://www.digikey.com/product-detail/ ... ND/1559519 (The size may be slightly off since I can't measure it on the board.) If you have a soldering iron and you're replacing the board anyway, you may as well give it a shot first. Nothing to lose at that point. Of course, if it's still under warranty then that's you're best bet, but it sounds like you've had this box for 6 years or so.

Rando
Posts: 757
Joined: Tue Jan 06, 2015 3:24 pm
Location: Boise, ID
Contact:

Re: Controller Box Problem

Post by Rando »

Kayvon

Well, that might be what's on that board...but it's also DEAD....

On the two units I have, the aluminum electrolytic can clearly says 100uF on it, and in-circuit (no power, but includes other components parallel to it) measures at 140uF.

What that burned out one is, I have no idea, and would never use that as the intended replacement. Given it blew out, and that it's different from my two boards on which the 100uf ones have not, it's more likely it was stuffed wrong. So yeah, when a component blows, and it's different, how exactly do you decide whether it was because the part was defective, versus the wrong part was intalled? You look at what's on a non-failing board, that's what :P . And then you test the failed board with a replacement part the same as the non-failing ones. If the new part doesn't fail, it was wrongly stuffed/specified. If the new part also fails, or fails in a different way, only then do you consider whether that part somehow represents a legitimate design change, or that the problem likes elsewhere and the burned out capacitor is just the weak link in the proverbial chain of failure.

Tantalum capacitors, for example (not implying that's what in his board) are notorious for leaking those vital smoke particles when they're installed backwards, or even when they're run slightly over voltage. Given we haven't heard of a rash of these old boards failing at C15, I'm going with that board, or maybe even that lot of boards, might have been stuffed incorrectly, and incoming inspection didn't catch it. Either that, or that one component part was defective and failed. Either way, it's appropriate to start from what works, not following down a path where you're not sure whether that one ever could have. The big problem, of course, is that we don't get current schematics with each unit shipped....unlike my 60 year-old furnace's temp-control box, which has a full schematic inside the lid. Tis the ghost of "No User Serviceable Parts Inside" come back to haunt us. :(

Well, that's my method and reasoning, for what it's worth :D.

Regards,

Thom
=====================================================
ThomR.com Creative tools and photographic art
A proud member of the Pacific Northwest CNC Club (now on Facebook)

Rando
Posts: 757
Joined: Tue Jan 06, 2015 3:24 pm
Location: Boise, ID
Contact:

Re: Controller Box Problem

Post by Rando »

Yeah, I removed the other, after a little more looking ;-) ....'cause i was wrong :oops:

Here's my board and the aluminum can. The outside rail is the power supply negative input, which makes sense from the other components.
2017-05-02_084724_399.jpg
The mark on AlElects is typically the outside can, which is the negative terminal. MLC and Tantalums indicate the positive terminal.

Either way, replacing it with a 22 or a 100 of appropriate working voltage shouldn't significantly affect the light load of just turning it on, since this only feeds the logic circuits, not the driver transistors.
=====================================================
ThomR.com Creative tools and photographic art
A proud member of the Pacific Northwest CNC Club (now on Facebook)

User avatar
Kayvon
Posts: 552
Joined: Tue Oct 21, 2014 11:46 pm

Re: Controller Box Problem

Post by Kayvon »

I had assumed design change rather than a bad component. Looking at your board pic, design change seems even more probable. The 22-35L is a tantalum and they probably thought better of it later. Without the schematic I can only guess to its use, but if it's just trying to stabilize the voltage for the electronics then the exact capacitance (22uF vs 100uF) shouldn't matter.

Finding schematics on consumer products is next to impossible nowadays. It's a shame.

Rando
Posts: 757
Joined: Tue Jan 06, 2015 3:24 pm
Location: Boise, ID
Contact:

Re: Controller Box Problem

Post by Rando »

Kayvon wrote:Finding schematics on consumer products is next to impossible nowadays. It's a shame.
Yup! And what's even scarier is that the schematics provided for so-called "open" or "Libre" hardware are often incomplete or incorrect. Using them for debugging is often a dog-chasing-its-tail exercise in futility. Might as well just design a new one myself than try and resolve the discrepancies.

Cheers, Kayvon! Awesome to be in the (virtual) company of someone familiar with the magical and mysterious ways of keeping capacitators and inducitificators from allowing those wily little green men to let loose their electrons in tiny bits of smoke. That's so often what people don't realize about electronics: you have to keep that smoke INSIDE the chips.... ;-)

Regards

Thom
=====================================================
ThomR.com Creative tools and photographic art
A proud member of the Pacific Northwest CNC Club (now on Facebook)

User avatar
Kayvon
Posts: 552
Joined: Tue Oct 21, 2014 11:46 pm

Re: Controller Box Problem

Post by Kayvon »

I'm a fellow sparky. It's why I always appreciate your technical posts, Thom. Also, your aluminum milling, which I'm still too timid to try.

peterchua96
Posts: 70
Joined: Fri Feb 11, 2011 8:42 pm

Re: Controller Box Problem

Post by peterchua96 »

Dear Friends,

Thank you so much for all your feedback.

I got a feedback from their Customer Support and I was referred to this alternative Controller Box:

http://shop.nextwaveautomation.com/shop ... itemid=185

Will this Controller Box shown in the link be compatible to the Shark Pro Plus Controller Box that I had?

Pete

Rando
Posts: 757
Joined: Tue Jan 06, 2015 3:24 pm
Location: Boise, ID
Contact:

Re: Controller Box Problem

Post by Rando »

Pete:

Well, I think there won't really be any other choice coming from Next Wave. I was doing a little research yesterday, and it looks like the microcontroller chip on the older boards hasn't been made since...2004. Yup, the spec from Freescale Semiconductor show it's been out of production since then. Note that it uses the same processor architecture (68000/68HC11) that was in the original Macintosh computers. Yes, the FIRST ones....So, if NWA can't get boards, it's likely because their supplier can't get chips, because all of the ones that exist are already soldered into boards, or dead. :(.

However, realize this opens additional possibilities. There are quite a few other CNC controllers out there now, some good, some bad, some really cool, and new ones being invented daily. While the "easiest" thing might be to just go with the NWA box, the forums seem to show some lingering software problems with that one. I suspect it's not any real "quality issue", but rather the normal cycle of there still being kinks worked out. I'm sure the original controllers had a similarly long period of some annoying thing not working right...like the fabled "bad usb cable" issue they recited for years. And at least with the NWA boxes, you won't be left on your own to debug the Open Source LinuxCNC or RepRap code. And yet, there are FAR more people working on the code of those two, than NWA's controller-board vendor could ever hire. Which means that the GCode support of those systems is significantly better. Whether that matters, of course, depends on whether you need those things, of course.

For example, the ncPod controllers lost support in Mach back in version 3, with still no support in Mach 4. Whereas many others are actively supported. The ncPod guy claims he's designed "hundreds" of CNC boards since then, but the Internet would apparently disagree, with nothing new since about 2004 from him or anyone using the term ncPod.

Bottom line is, like with most things, how much is it worth to have a place to just send the box to when it dies, versus the effort and learning of a new system and not being tied to what amounts to a proprietary system? Personally, and granted I'm not normal ;-), I'd take the opportunity to ditch the ncPod-based controllers from NWA and move to a standards-based, open-source platform. Technologically speaking, ncPod is a dead-end product.

Regards

Thom
=====================================================
ThomR.com Creative tools and photographic art
A proud member of the Pacific Northwest CNC Club (now on Facebook)

Post Reply