Lorlin Impact Test System Disk...

On Sunday (03/06/2011 at 06:06PM +0000), Tony Duell wrote: What a great topic... as I have been doing this exact effort for the past several weeks. I have completely cleaned and brought back online two RL02 and a third is just entering the process. Before I ever try to spin one up, I take the front panel off and remove the black pre-filter. It is guaranteed to be disintegrating and turning into a fine black powder. Exactly the kind of thing you don't want sucked into the drive. I vacuum out entire front area behind the front panel and the backside of the panel itself. I then take off the top covers and vacuum out the well where the pack sits and wipe the entire area down with a damp cloth. Clean out ALL the junk... cob webs, hairs, there will surely be a lot of junk in there. Then I clean out the rear near the fan. The fan blades will be loaded up with dust and I take off the grill and wipe those down with a Swiffer duster cloth as well as vacuum the whole fan assembly. Then there is the issue of the "absolute filter" which is the roughly 3" x 9" HEPA-like filter accessible from the front. I was fortunate to have one of these that was nearly new but others have been polluted by the disintegrating black foam pre-filter. I am currently looking at one that is completely black on the intake side-- totally loaded up with the black carbon-like dust that the pre-filter turns into. Not cool. I was just about to post to the list asking if there is a source for these absolute filters still (which I doubt) or if anyone has a process for rebuilding them? I've already stripped the filter element out of one and so I have an empty frame and am then considering modifying a HEPA filter for a room air cleaner or other appliance that uses a HEPA filter module. I will have to cut the filter to size and then glue it into the frame (similar to how these originals were made) using an elastomeric caulk or similar. But-- is there a better way?? After dust and dirt cleaning... then I clean the heads. I very carefully use a cotton swap soaked in isopropyl alchohol... 91%... drug store stuff... and rub it between the lower and upper head, sort of twirling it back and forth which removes the junk on the head pretty easily. I do this several times until the cotton swap comes out clean. At this point, I am brave enough to try to spin the drive up. I will take a known good not not precious pack and put it in the drive. You of course have to have a controller connected to the drive and powered on so that the drive gets clock over the interface. At this point you should have the fault light out and the load light lit. You can press load and hear the drive spin up. Hopefully. I have one drive that will spin up, attempt to load the heads and then immediately fault. After several experiments trying to get this drive to load the heads so that I could do some of the alignment procedures, I found that it was actually crashing the head into the media and I wrecked that pack. That drive is now a parts donor. I don't know exactly the failure mode but I suspect that the spindle is no longer aligned with the heads... due to some abuse like the thing being dropped. Dunno for sure. If you get to this point and the drive spins up and doesn't crash the heads then you can try to read the pack using one of the diags in XXDP2.5. I use ZRLMB1 which will read the entire pack and report the bad sectors found. On one of my drives I had consistent failures at cylinder 490 and beyond... which after performing the read amplifier gain adjustment procedure found here, http://bitsavers.org/pdf/dec/disc/rl01_rl02/EK-RL122-TM-001_techAug82.pdf the problem went away. I set the gain just slightly hotter than they spec on the theory that many of these packs are old now and probably produce a weaker signal than originally designed. I repeat the test across a selection of packs to get the average signal level too since I don't have a "standard" test pack (saw one on eBay for $1000 the other day!!) So, I have had good luck with 2 out of 3 RL02 so far. These filters might be an interesting challenge but I don't think insurmountable. Chris -- Chris Elmquist