My cnc computer boots just fine, then about 30 seconds later it kicks out like a circuit breaker trips. But there is no cirucit breaker that I know of.
Sould i just replace the power supply? Other possible trouble?
Man its HARD to live without CNC. I have to turn a crank to make a part, it sucks.
Could be many things, but check the simple and ceap things first: Check that the CPU cooler has not become dislodged. 30 secs is about the time it takes for the CPU to overheat.
Bad muffin fan in the power supply, or on the CPU heatsink. If the bearings are wearing out the might squeal for a while before they quit totally. And it might be slow to start turning.
Someone might have put the CPU Cooler heat sink on dry, without the white heat transfer grease. Or removed it to clean and wiped it dry, without re-gooping it. It's there for a reason...
Get a Power Supply Checker if it's a modern PC style unit, plugs into the power supply leads directly - it'll show green lights for all the
+5V, +12V and -12V lines. And if it's green for thirty seconds and then one line goes Red, then it's for absolutely certain.
If it has a 'Soft' power switch that turns on the power supply remotely like on a PC, it could be a bad power switch.
It could also have separate Industrial style supplies for 5V and 12V, especially if they run the stepper motors from the same source, and it has to be the same source feeding the driver boards - it would take a lot more current than a regular PC supply. Then you have to get in there and clip on a voltmeter, and see what happens after the magical thirty seconds.
You would have to make an adapter from the old plugs to the new style. I may have a good 20 pin plug, but I will have to go through a pile of power supplies to look for one.
Is it built on a PC, or is it a commercial CNC machine? I think that you converted it, so it is probably built on a PC of some flavor.
In that case, the power supply is certainly one thing to look at. Another is all the fans. If certain fans are not running, temperature sensors get too hot and it shuts down to protect itself.
In particular, given the speed of shutdown, check the fan for the CPU. And if you have replaced the fan/heat-sink assembly, what kind of thermal connector is there? In the old days, it would have been a white grease (heat sink compound) full of some thermally conductive compound in silicone grease. That should be scraped off both parts and fresh compound put on.
In later times, it is a silicone rubber pad with thermal conductors embedded throughout it. Sometimes these don't connect as well for a second use, so they should be replaced.
At least you *have* a crank. My Bridgport which used to be a BOSS-3 Series I had no cranks except for the knee -- and the X-axis leadscrew does not even rotate -- instead the ball nut rotates around it, where you can't get in and connect a crank. :-)
Other possible problems would include electrolytic capacitors which would overhest and short -- some in the power supply, and some on the motherboard.
If you have a spare power supply swap it in and see what happens. First look at all the fans when you first turn it on. They should all go at full speed to start, until the motherboard decides that it is cool enough and slows some down.
And look at electrolytic caps (usually metal cans on end) for bulging in the end -- with or without splits at the intentional weak points put in as safety fuses, so you don't have the can exploding.
This is the kind of problem we'd troubleshoot with a hair dryer and a can of freon. A paper and masking tap box around a chip, board, or section of a board can isolate a circuit right quick.
But these days a digital non-contact thermometer might be faster.
They didn't need the "power-good" so, it was even easier to use a DVM. You didn't have to short two pins to fire it up. I still have a couple of those kicking around.
Well Duh! But its slow! Ive had several of them blow the -12vt line or the -5vt line...and those can be a pain in the ass to resolve when there is a problem.
Don't really work in the shop in "farm" season, so i waited for a power supply and supply volt tester.
Turns out a plastic ear had broke off on the CPU cooler hold down clamp so the CPU quickly overheated.
Had to drill and tap out a mounting spot on the motherboard to 8-32, no 6-32 to be found here, then drill a hole in the hold down clamp. Installed an all thread from motherboard through the clmap and tightened.
All is well. I think this may be the first time I fabricated anything for the inside of the computer.
Careful drilling holes in those boards, they are multi-layer and tight spacing so what looks like a clear area on the top and bottom layers may have critical traces on the layers in-between. A better option is to CA/epoxy a mounting point to the surface of the board, or to fabricate a metal replacement for the broken plastic part.
I fabricated EVERYTHING inside and out on my first computer, a wire-wrapped 8080.
The crudest fab work I've done inside one was hacking down an oversized sheetmetal flange on a replacement Apollo Workstation power supply with my home-made survival saw/knife. I switched hats from being a DRAM memory controller ASIC designer to a medieval smith ripping off excess metal with a hand tool. jsw
"Jim Wilkins" fired this volley in news:kpau73$26l $ snipped-for-privacy@dont-email.me:
Yeppers! My first one was built in a wooden chassis. Mine was wire- wrapped, too *but I got lucky, and found one of those Vector 'self- stripping' wire wrap tools that fed from a spool! (oh, yeah!)*. It was built on Augat wire-wrap/socket strip panels I got for cheap as surplus from the 'cape'. All solid gold, everywhere.
While my first, second and third computers were built from kits (Altair 680b, Technico 9900, and SWTP 6800), I built various dedicated controllers on wire-wrap boards, and a replacement CPU card for the SWTP
6800, to allow getting the full speed out of a 6800B (2 MHz). SWTP fed the baud rate clocks down the bus, and drove them from dividing down a crystal which also determined the CPU clock rate -- something a bit North of 800 KHz, IIRC. Divorcing the baud rates from the CPU allowed other benefits as well. After a while, I moved to the SWTP 6809 and an interesting OS -- OS-9 (*not* the much later one from Apple for the Mac by the same name. :-) This was sort of a unix-like OS which could run almost totally in ROM if you so desired -- just enough RAM needed for stack space and data.
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.