hull@hao.ucar.edu (Howard Hull) (08/09/88)
Ah, I love those CBM flames. So warm, so toasty. But this is August... Naw, I just put out articles like that to get the CBM folks to come alive. A couple of whiles ago I put out one on how to adjust a 1080 monitor. George Robbins responded with a beauty of a retort. He's a serious hardware designer with capacity one can really learn to appreciate. I LOVE that Amiga hardware. However, It looks to me as though Amiga monitors are something else. That's why I'm really looking forward to the Hedley monitor. Looks like this baby has been taken care of from concepcion... I just wanted (like the original poster) to see what you guys were thinking. Nice response. And yeah, you're right. You gotta be careful of what I tell you. I talk possibilities. All possibilities. I don't earn my money from doing stuff. I get it for hooking everyone else into getting stuff done while trying to bash my brains out. :-) After all, jpc DID ask once and got no takers. So I set you guys up. Congratulations, you bit. I love it. Now back to Amiga monitors in general. I remember a spate of articles way back when (musta been August '86 through February of '87) where people were talking about Amiga 1080 monitor arcing. Nobody had anything reasonable, either for suggestions or real data. And CBM never said peep, even with some goading from the troops. Finally there was an article by one guy who recommended corona doping everything. Stupidly, I lost the article, so I can't quote from it. I remember going by my Amiga/IBM/Mac dealer's and seeing their Amiga monitor sitting there with the hand and the disk brightly displayed (they never got to where they put the Kickstart disk in the thing), while it was going "SNICK" and rolling over in vertical every minute or so. They never fixed it. In May of '87 they quietly dropped the Amiga line, and for a month after that, I tried everything I could think of to buy their A1000 for something less than retail. They never budged. (What I really wanted was their dealer demo disks. I later got them through another guy who almost became a dealer, but then bailed and sold me the demo machine with LOTS of software (including the dealer stuff) for a kilobuck.) Annnnnyway, last winter, my 1080 started SNICKING. I have all the documents CBM would sell me with respect to the A1000, but nowhere in the lot is there anything closely resembling a schematic for the 1080. My 1080 is full of ingenious Japanese MSI parts, which of course are unfathomable without data sheets. This same 1080 has also recently developed an audio intermittent, one which behaves suspiciously like either a reversed biased electrolytic capacitor syndrome (especially manifested when I run SubLOGIC's wonderful Flight Simulator - audio briefly shorts out now and then), or like the vibration-cracked power IC solder joint syndrome (commonly found in DEC VT100 horizontal driver circuits). What happens is that the power transistor junction undergoes oscillating stress from the applied electric fields, and actually deforms the silicon die (verrrry slightly). The compliance numbers must be best described in microns per ton per square inch. This stress then propagates sound waves out on the IC leads, flexes the solder, and eventually cracks the solder joint. It's easily fixed (for a while) by resoldering the power transistor leads. The key to avoiding this problem is to make sure that the IC goes into the eyelets with a class 5 fit (interference) so as to drive the square lead edges into the copper in a gas-tight seal. The solder is then added to encapsulate the joint. The 1080 horizontal drive, I must say, has been rock solid in this respect, so somebody must have learned something somewhere. However, when I went (casually) to hunt for the audio power transistor, I ended up wandering around in the right front part of the PC board where there were a couple of those mysterious Japanese MSI parts. End Game. Moving on to the monitor arcing problem, I invoked my superior powers of observation, and came to the conclusion that the vertical sync rolls JUST BEFORE the monitor goes SNICK. This means to me that she missed a vertical sync pulse. There have been a couple of other net articles that have come startlingly close to saying almost the same thing. Now when I have a problem cropping up, suddenly or otherwise, the thing I always ask is "OK, guys, what's changed? In my case, there has been only one significant change in my hardware prior to the onset of the problem. The change was that I added a Michigan insider. I have not bothered to do the PROM ground thing yet (criminally lazy, as I am). The other thing was that winter happened along (87/88). Now that summer is here, the SNICK problem went away. I expect it will return in December. Now then, someone on the net recently suggested that monitor arcing could be the result of high humidity. I wouldn't challenge that, particularly if the arcing happens before the sync rolls. But here in Colorado, the winters are exceptionally DRY and COLD. More likely in my case the symptoms point to a missed sync pulse (and consequent vertical output amplifier saturation due to the continued ramp-up of the un-reset vertical drive current to the point that it either hauls down the power supply, killing the H and V drive pre-amps, or zots some MSI halted-beam-detect-protect-trip MSI gizzy whizzy, which then crowbars the drive to keep the beam from burning a hole in the phosphor/glass). By now I have confiscated the wordiness-star award. But continued reports of monitor arcing in Amiga monitors would indicate to me that by and large the Amiga engineers do not directly design them (they must supervise a contractual deal with the Far East). If they did, they would be done once, and done right, like the Amiga 1000. Last week I picked up an Amiga 2000 to augment my two A1000's, and I am so far real happy with it, too (even though I am now perhaps waiting for the monitor to show me something I didn't really want to see). SO, the upshot of all this is: WHEN IS CBM GOING TO START INCLUDING MONITOR DOCUMENTATION WITH THE TECHNICAL DATA PACKAGE so that we can see how hopless the situation really is? I am DEAD RAT SURE that there are failed-trigger caveats in the design of all the monitors CBM has whipped up from our Far East friends (are you SURE we won WWII?). I think that my difficulty in particular is in the A1000/A1080 RGB interface, due to marginal signal levels there, and is caused by a cool-temperature-related design problem coupled with slightly lower supply voltages (courtesy of the Michigan Insider), and that the 1080/ A1000 design is marginal in that one respect. Worse yet, there must be a similar problem in the A500/1084, given the number of reports from A500 owners. Here the common thread is the narrow design margins for the A500 power supply. To test this, last winter I ran my 1080 for a week on the NTSC interface [ugh!] and narry suffered a blip, let alone a SNICK! So maybe you'd like to tell me a little more about what I know that ain't so on this one, oh humble servant of the computer masses... Howard Hull hull@hao.ucar.edu "'Taint what we know that gets us in trouble. It's what we know that aint so." Will Rogers the 1st. "I'm going to have to live with this for the rest of my life." Will Rogers the 3rd. Aren't we all, seems as how we put up the money for those things. So, we should be surprised if they are on occasion inclined to use them?
lel@wuphys.UUCP (Lyle E. Levine) (08/10/88)
Hmmm... I've been hearing lots of complaints here about Commodore monitors. Well, my 1080 works perfectly and looks MUCH better than those new 1084s. Tell you what. I would like to upgrade to a Multisync FlickerFixer combo but still have my 1080. Anyone wnat to buy it? I have never had a pop, glitch, wiggly window, jumping text or any other weird manifestation I keep hearing about. Please email any requests. ========== IBM is a Division of Sirius Cybernetics Corporation "their fundamental design flaws are completely hidden by their superficial design flaws." - "So Long And Thanks For All The Fish" Lyle Levine: Paths -> ihnp4!wuphys!lel Best way: (314)889-6379 uunet!wucs!wuphys!lel
bryce@cbmvax.UUCP (Bryce Nesbitt) (08/25/88)
In article <553@ncar.ucar.edu> hull@hao.UCAR.EDU (Howard Hull) writes: | |Ah, I love those CBM flames. So warm, so toasty. But this is August... | .... |SO, the upshot of all this is: WHEN IS CBM GOING TO START INCLUDING MONITOR |DOCUMENTATION WITH THE TECHNICAL DATA PACKAGE so that we can see how hopless |the situation really is? The 1080/X schematics can be purchased from the Commdore repair/service folks. Call 'em and find out all the details. You may need to be a certified service center or something, in which case bribe some technician at an Amiga dealership.