Re: Meridian Flip
|
Gosh, Rob! I have had the same experience as you have... load a "known good model" and ... very weird... the mount will slew to point my scope to the floor while I wonder what button to push to stop it! I don't know why the system would do this, but there are many possible causes.? Any of the wrong values of time/date/longitude... that data could be corrupt.? Another reason is that the pointing model has something totally whacked out. ? The thing that has worked for me to recover normal operation is to completely zero out the T model parameters. ? Does your Cold Boot send you into limbo?? Then it must be a wrong UT, latitude, longitude, date, time.? The zero model values should still get you relatively close to your first alignment star (maybe an hour off in RA if the Daylight Savings time correction gets it wrong.) Then I do what you do too: I use Stellarium (with Stellarium-scope) to pick my 3 stars on each side of the meridian.? I don't usually use the hand controller list except for perhaps the very first very bright star, because I don't know what will be on that list, and my house surroundings limit my sky view.? I am limited to mostly overhead and about 90 degrees E-W region and 120 degrees N-S.? So I must pick my alignment stars from a sky chart. ? Like I said, I found it too easy for the PC to send in its default model parameters and mess me up.? Be careful that is not happening to your system. ? Now you have a "bad" sky model in your Gemini.? That has a value..it has clues to the problem.? Read those values back into a file...maybe write them down too.? Then try to make a new good model that is working well.? Compare the model values and try to figure out what might be a bad value. ? All we can say for sure is that it is certainly a Gemini problem... the question is what setting has gone bad. Can it be hardware/electronics and not a model value? There have been sparse reports that an EPROM chip in the Gemini-1 has gone bad.? Though I never would have believed it, I have measured some of these chips to go bad.? Usually it is one bit gone bad, but many can also go wrong. ? In that case, the firmware is corrupt and the part should be replaced.? I have now bought an EPROM eraser and can test and reprogram these chips for anyone with a questionable one....or I have new ones too ready to send to anyone (the last version Level 4 v1.05). ? Can an SRAM go bad? Yes... but the Gemini-1 is supposed to test the SRAM for validity...and show a dot and not boot up if that SRAM is bad.? If the system boots, it is likely the SRAM is OK. ?(and I have replaced a bad SRAM for one unit...). ? Can an SRAM glitch? That's an open question.? The T-model values inside the Gemini-1 are held in SRAM which is powered by the 3V battery.? That battery circuit has a Schottky diode after the battery, so the voltage coming off the Schottky diode is about 2.8V to the SRAM.? Likewise the voltage after powerup with DC external applied (12V-18V) comes thru a 5V regulator, so after that goes through its Schottky diode, only about 4.7V comes out to the SRAM. ? The SRAM itself does not care as long as its Vcc is over 2V, it should hold its data.? But the pins of the SRAM are getting the full 5V from the other chips (and microprocessor) that it talks to, and the spec on the SRAM is that its pins should not see more than 0.5V difference... not more than 0.5V over its Vcc, or 0.5V below ground...else the pin can get damaged.? I think (without proof) that this is why some SRAM chips go bad yearly, for someone in our group. ? - All the best, and when you find the culprit, let us know!!! Michael On Tue, Aug 15, 2017 at 12:02 PM, sactowriter@... [Losmandy_users] <Losmandy_users@yahoogroups.
--
|
Michael Herman