• Please review our updated Terms and Rules here

Fried 486 processor socket

D

DamienC

Experienced Member
Joined
Aug 2, 2010
Messages
72
Location
Southern NJ USA
Have a question about a potentially dead 486 motherboard. Kind of a long story though...

I picked up a nice late model 486 motherboard on eBay, a Lucky Star LS-486E. Has some nice features: Onboard I/O controller, PCI slots, 72-pin ram slots, coin-style CMOS battery, etc.

The auction said the board was tested and working and being sold as-is. Like with all eBay things I took this with a grain of salt, and figured it was cheap enough to risk purchasing.

So the board came in yesterday, and was supposed to come with a Cyrix DX2-66 CPU and 8mb RAM. I gave it a quick visual inspection and blew it out with some compressed air because it was dusty. I didn't notice anything immediately wrong with it, so I plugged it up and powered it on.

After roughly two seconds, I heard a large cracking noise and smelled something burning, so I pulled the plug. I took the motherboard back out, and sure enough, the CPU was fried.

The processor socket was a nice ZIF socket with a lever, so I popped the CPU out and inspected it. One of the pins on the CPU was definitely fried. When I removed its heatsink, the chip literally snapped apart like a saltine. Also, the chip was not a Cyrix DX2-66 as advertised, but a "It's ST" 486DX2-80. Either the chip was oriented wrong, or the wrong voltage was set on the motherboard, or both. In any case, the CPU is definitely dead.

As for the board itself, the one hole in the CPU socket where the fried pin was is now blackened. I did a very close inspection of the rest of the board and could find no obvious physical damage. I'm wondering if the motherboard is still usable. I unfortunately don't have any other 486 CPUs lying around to test it.

Anyone have any suggestions? I'm leaning towards the board being unusable.
 
I would say find another 486 ship, check your settings and orientation and try again.
I blew up a DX 66 the same way by installing it the wrong way. The chip was dead and the ZIF slot had a blackened mark but the board still chugs along today with a 5x86.
 
Here's the problem as I see it. If the fault is in the motherboard (e.g. bad regulator, shorted lead) you're just going to fry another CPU. If you can check the CPU voltages to make sure that they're not out in left field, then you may want to try another CPU (or at least a 5V CPU).

As far as the ZIF socket goes, you can usually pry the top part (the part that slides) off the bottom (fixed part) so you can inspect the contacts. If the lower part is damaged, you're best off tossing the motherboard rather than trying to replace the socket.
 
The ST 486s are relabeled Cyrix CPUs. ST was one of the fabricators Cyrix used, and as part of the deal, ST got some of the chips. Same arrangement as the IBM/Cyrix arrangement in the 6x86 days.

Now, as for the board... I put the CPU in wrong once. It was a very bad day. In my case, I smoked the motherboard, not the CPU. In yours, if the CPU was in the wrong way, it's possible that the CPU smoked before the motherboard could be damaged. Or both could be gone. Unfortunately 486s had no key pin, so you could put them in wrong and blow stuff up. :( That was the worst thing there was about the 486.

As I recall, the Cyrix and ST 486s were 5v chips, like Intel.
 
Check the specific Cyrix model. 486DX2-80 came in a 5v model but the 486DX2-v80 was also made in 3.45v and 4v. Cyrix's own FAQs show DX2 as having a 3v model as well. Twill be quite confusing. All of which assumes the chip label wasn't modified to indicate the rarer and more valuable variant. Though unless I can find clear evidence of the wrong voltage being applied to the CPU, I would junk the motherboard.

http://www.cpu-collection.de/?tn=0&l0=co&l1=Cyrix&l2=Cx486+DX
 
I would bet that the motherboard lived, and the fault was the CPU orientation or compatibility. I'd do testing with a common chip, though, DX25, DX33, DX266, SX25, SX33... no need to ruin an uncommon chip.
 
Yeah, think I'll try to find a cheap CPU and see if it still works. Too bad about the DX2-80, I never had one at that odd speed before and was kind of curious about it's performance.

It wasn't all that expensive so I'm not too chuffed about it. :p
 
Think that's bad? I fried (not totally dead, but unusably weird-acting) an IBM-branded Cyrix-made DX4-100 chip with no heatsink/spreader necessary by putting it in something that didn't provide the proper voltage. I also fried (completely) a Cyrix 5x86 (proper one) 120Mhz by putting it in a motherboard that didn't provide the proper voltage as well. At least yours was an honest mistake and not the highest end chips.. The IBM one I was a noob, and the Cyrix I was just stupid with..

Damn well learned my lesson about voltages, though. On the bright side I discovered that Am5x86 chips can survive 5v for months without ill effects even though they're designed for 3.45v..
 
The AMD DX2-80 ran faster than the Cyrix DX2-80 did. Cyrix's true 486s, while much better than the SLC/DLC trickery they did early on, still didn't run as fast as Intel's or AMD's. A Cyrix at 80 will outrun an Intel at 66, but the difference between an AMD and a Cyrix at the same speed was noticeable. Performance of both was just like it suggests, somewhere between a DX2-66 and DX4-100.

I used a few Cyrix 486s in my day, even though performance wasn't quite as good. They were less expensive than even AMD 486s. As I recall, a Cyrix DX2-66 was considerably less expensive than a 50 MHz Intel, and it would outrun the slower Intel chip. And since most of the Cyrix chips ran at 5 volts, they were a cheap upgrade for older 486s. They ran $25-$35, as opposed to $150 for an Intel Overdrive CPU in a pretty retail box.
 
To my complete surprise, the board still works!

I got a cheap old DX-33 CPU to test the board with about a week ago, and it didn't seem to work. I figured the board was fried and that was that.

This weekend, while cleaning up my computer room I decided to give it another shot. Figuring I had nothing to lose, I decided to start playing around with the bus speed jumpers. Lo and behold, after a few tries the DX-33 CPU booted. It turns out the bus speed jumpers listed in the manual were labeled backwards! I had set the bus speed for 33MHz according to the manual, when on the board itself it was actually 40MHz so of course it didn't work.

I installed a Pentium Overdrive 83MHz CPU and now it's running great. Here's hoping I never have to open the case of this one for a while. :p
 
You must log in or register to reply here.
Back
Top