Tokyo PC Users Group
	  Home Page
Home
Members Only
Newsletter
Newsgroups
Become a Member
Meeting Info & Map
Officers
Members
Corporate Members
Photos
Workshops & Training
Other Clubs
Job Hunting?
Constitution

From the Newsgroups

I'm usually reluctant to reprint extensive material from the newsgroups in the newsletter, for several reasons including my assumption that many/most of the AJ readers will already have seen it. However, the thread that I've excerpted and quoted below started, as you can see, with a comment from the TPC's resident guru and all-round great guy Jim Tittsler, directly referring to an article by Ralph Sumner that we've been printing as a series. Since the pressures of (paying) work prevented Ralph from getting the next installment in, I thought you might find this interesting to read in its place. I certainly learned a thing or two from reading the various comments. I'm hoping to be able to continue the "Roll Your Own" series from next issue or so. - Mike]


From: Jim Tittsler

Ralph Sumner is conducting a great series of articles about how to "Roll Your Own!" PC from parts in the Algorithmica Japonica. He describes the considerations behind choosing to assemble your own machine and then the mechanics of doing so.

However there is one paragraph in the July installment I would take issue with:

  [Ralph describes the risk static can pose to small feature electronics. He then continues...]

  Do you have one of those neat wrist bands with an alligator clip cord that connects to a ground, to keep you free of static electricity while you're touching the parts? If not, you can make your own just by tying one end of a wire to your wrist and the other to some sort of metal ground (making sure you've got bare wire contacting both your wrist and the ground).

Please don't do this. While you might reduce the chance of static damage to the component, you expose yourself to an unnecessary shock risk. The "neat wrist bands" are designed to have a *very* high resistance between the operator and ground. These 100s of kilohms or a megaohm are still low enough to bleed off the thousands of volts of static charge that can build up on you as you pet your nylon cat, but high enough that it isn't a hazard if you were to somehow come in contact with an AC line. You could make your own grounding clip by using the bare wire idea and attaching a one megohm resistor between the wire and the ground point (at the ground end obviously). The lazy technician's way of working without a strap is to touch the chassis with the other hand away from sensitive areas while inserting boards and components in the hope that while both you and the chassis could be at some elevated potential, it will be about the same and it won't be flowing from/to you through the delicate component.

In addition to Ralph's comment that avoiding carpeting can reduce static risk, a hard floor also has advantages in finding small parts that you or the cat drop.

  Oh, and don't use that magnetic screwdriver.

There is no reason not to. It presents no static or magnetic risk to your machine. And I think you will find it much more convenient for holding the screw while you are getting it started (or fishing it out of where you dropped it with the non-magnetic driver :-).


From: Charles Lipton

Agreed. I have been building my own boxes for years and the magnetic screwdriver is one of the best time savers going. It is also one of the best ways to get around sausage-sized finger problems when you work on the very small NLX boxes ... but they work well for all. And, as Jim says, they find dropped metal parts better than your eyes do. I keep an automotive mechanics magnet on a stick close by for chasing up the parts that get feet on them and run away when dropped.

Just don't stand them up inside the box you're working on or rest them on the mother board.


From: A. Sajjad Zaidi

I agree. It would probably be safer to not use any grounding than to do this. I've built many many systems and have repaired quite a few as well, but never had to use any grounding other than touching the case occasionally.

Maybe I was extremely lucky, but I dont think this much grounding is required. I do know of certain people (d are you reading this?) who damaged video cards or motherboards by working with them after being charged from silk or other such materials, but tha'ts if you're very careless...<snip>...A few people have expressed concern on magnetic drivers and you're right, they're safe and convenient to use.

I would like to point out some other mistakes that can be made. More than once, I've unwittingly plugged in a part into a running system (amazing how quiet a system can be without a hdd or other such noisy devices). Was lucky enough to even get away with RAM chips, but don't count on it.

A common mistake is attaching a floppy drive (if you're old fashioned enough to use one :-)) data cable the wrong way. Its hard to tell, even for seasoned builders. No permanent damage with this mistake though.

One last stupid mistake I've seen (but never did myself) was putting the floppy power the other way. This is tough to get wrong, but it will KILL your drive.


From: Sam Julien

Another grounding note is that if your power cable is properly grounded (3 pin) to the wall socket, then leave it in the case when fiddling (making sure the power is off...), as the cable is still grounded....

Think I read it on Tomshardware.com (think that's the url).


From: Kevin Sullivan

Floppy data cable the wrong way on bootup gives a steady green light on the floppy.

Don't think you can hot-swap keyboards, I've given the coup-de-grace to two of my oldest motherboards (486DX33 and PPro) this way. On the other hand, there are always interesting new motherboards with much better performance out there...


From: A. Sajjad Zaidi

Must be an issue with older motherboards. I hot-swap keyboards very often, both at the office and at data centers. To be fair, these are mostly server class machines. Have done the same with mouses and it works great unless you're running NT.


From: Sam Julien

Hotswapping anything with WIN sux.....


From: Roland Hechtenberg

I have rather different experiences, and except for parts inside the box, I can hot-swap practically all external units, thanks to USB, Firewire, and reinitialization of the SCSI card without rebooting.


From: A. Sajjad Zaidi

As far as I know, not all SCSI cards support hot-swapping. Is that true? I know most 'Adaptec AIC-7xxx' support it, but is it the same for others?


From: Roland Hechtenberg

I don't know whether the SCSI board supports it or not, but Windows does. Naturally, I switch off the power to the unit being swapped, but after connecting a new unit to the chain, I go to Control Panel, System, Device Manager, select the Properties of the SCSI card, and update the driver, and then the newly connected unit is recognized without rebooting. This definitely is a feature of Windows and not of the SCSI card.


[And the discussion continues into various issues of stability and SCSI functionality. Visit the tpc.hardware newsgroup to see the rest. - Mike]


© Algorithmica Japonica Copyright Notice: Copyright of material rests with the individual author. Articles may be reprinted by other user groups if the author and original publication are credited. Any other reproduction or use of material herein is prohibited without prior written permission from TPC. The mention of names of products without indication of Trademark or Registered Trademark status in no way implies that these products are not so protected by law.

Algorithmica Japonica

August, 2001

The Newsletter of the Tokyo PC Users Group

Submissions : Editor


Tokyo PC Users Group, Post Office Box 103, Shibuya-Ku, Tokyo 150-8691, JAPAN