|
Folks
I have successfully run a PemPro 3 session - 12 worm cycles - created a good PEC curve and am now trying to load it to the Gemini controller but keep getting serial port comms errors and serial port timeouts.? It looks like PemPro is trying to write the curve data out too fast?? Any ideas?? Gemini log file attached...
Regards
Mark
|
On Fri, Mar 22, 2024 at 07:14 PM, madastro wrote:
Folks
I have successfully run a PemPro 3 session - 12 worm cycles - created a good PEC curve and am now trying to load it to the Gemini controller but keep getting serial port comms errors and serial port timeouts.? It looks like PemPro is trying to write the curve data out too fast?? Any ideas?? Gemini log file attached...
Regards
Mark
Hi Mark, you are getting a lot of communications errors in the log, long before you try to program PEC. You may need to replace the serial cable, or the USB->Serial adapter, or both. Regards, ? ?-Paul
|
Paul
Many thanks! ?I have been through a suite of tests. ?I have tried 3 cables and 3 usb convertors - tried all combinations and then made up a new cable - nice and short RS232 to db9. ?My pc has a db9 connector as one of the options. ?I also have two Gemini 1 L4 control panels - so I swapped the panels and got the same results. ?Am at a bit of a loss¡.
Regards
Mark
|
Hi Mark,
Hmm, the only thing I can suggest, since most of the errors did get recovered by a retry, is to increase the delay between commands when programming PEC from PEMPro. If I recall, it's just how many milliseconds to wait between commands. Make it some reasonable number, not too large, otherwise you might be waiting for a long time for the whole PEC table to be programmed. 5-20msec should be plenty.
Regards,
? ? -Paul
|
Paul
Thanks - I tried setting that all the way to 100 (the top end) and no difference. ?
Regards?
Mark
|
On Mon, Mar 25, 2024 at 04:25 AM, madastro wrote:
Thanks - I tried setting that all the way to 100 (the top end) and no difference. ?
Mark, can you please upload a log with the errors with 20ms delay? Just want to see if there's any patterns I can pick out between the two logs.
|
Paul, Michael I waited for the Aten cable to arrive and ran these tests just now with the new cable.? Unfortunately no different.? Attached are:
- Trouble shooting - gemini_status log
- Trouble shooting - ascom.gemini telescope.net log
- Gemini log file run at 20ms delay as requested.
I also checked the port settings for com9 - I notice the flow control is set to 'None' - that would be a major issue if the gemini/pc handshake were capable of hardware/software flow control?? But with the pins used at the DB9 end, my assumption is that there is no hardware flow control...  Regards Mark
|
Hi Mark,
The problems you are having have one thing in common: the computer. When using a USB to serial converter, have you tried using a different USB port on your computer? USB ports on a computer are usually designed in pairs, so the USB port next to the one you are using may be the problem, especially if the neighboring USB port is to a high-bandwidth device like a camera. So, try using a different USB port.
Also, you want to keep the USB cable as short as possible. The max recommended USB cable length is only about 15 feet, but RS232 cables can be much longer, up to about 300 feet. I've found that USB/serial converters with a short USB cable to the DB9 end typically work best because you can build/use a longer RS232 cable between the converter's DB9 to the mount.
-Ray Gralak
(Author of PEMPro)
toggle quoted message
Show quoted text
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of madastro
Sent: Sunday, March 24, 2024 2:18 PM
To: [email protected]
Subject: Re: [Gemini_users_io] Gemini 1 L4 Issues loading PEC Curve from PemPro 3
Paul
Many thanks! I have been through a suite of tests. I have tried 3 cables and 3 usb convertors - tried all
combinations and then made up a new cable - nice and short RS232 to db9. My pc has a db9 connector as one
of the options. I also have two Gemini 1 L4 control panels - so I swapped the panels and got the same results.
Am at a bit of a loss¡.
Regards
Mark
|
Mark. ?I¡¯m tempted to send you one of my GEMINI BT devices. ? I¡¯ve had not trouble connecting it at 56k and even higher 115kbps I think?
It¡¯s just a BT RS232 Serial Port Protocol. ? I Wonder if this would help. ?But FTDI etc Rs232 converters are capable of these speeds no problem nor does it appear a connection data rate issue?
Which reminds me have you tried connecting at higher data rates to see if this allows better flow? ? I¡¯m sure you have.?
FWIW really odd!
--
Brendan
|
Oh I have a new FTDI USB/RS232 G1 leads I can loan you.?
--
Brendan
|
I noticed what looks to me like an odd valued (arbitrary)TOD difference between computer and Gemini in your Ascom logs.?
Not sure if this would impact what you're trying to do,? but wanted to be sure you're aware of it. The comms on com9 in the second log look normal, not knowing what you are doing at any particular moment. It didn't look like a cable or transceiver issue.?
I'd fix the time sync and try whatever your doing again as it might cause unusual problems.
Just my 2c.
|
Ray
Yes - tried all available ports.? Serial cable (DB9 to RS232 4p4c) is 2 foot long, untwisted pair cable but also unshielded.? The FTDI to USB are all very short too (I have 3 of them, plus now an ATEN version.? Thanks fo the advice - but still perplexed!? Tomorrow I have the morning clear will will try a completely separate PC with just the gemini and pempro software installed....
Regards
Mark
|
Brendan
Actually - good point.? I have tried to slow it down but connections reverts back to 9600.? I'll try that...? ?Have 3 FTDI usb to serial cables, all new'ish so thanks for the offer...
|
Oh - well spotted - I had missed that.? I run Dimension 4 to keep the PC time real just in case I do any exo planet measurements - it may have caused the issue as it impact the PC time but not the Gemini time.? I haven't done any exoplanet stuff for ages so can jsut disable that from running and resyn the times.
|
Tried correcting the time but no change - and I have now disabled Dimension 4...
|
On Fri, Mar 29, 2024 at 07:33 AM, madastro wrote:
Tried correcting the time but no change - and I have now disabled Dimension 4...
Mark, thanks for uploading the new logs. I'll take a look over the weekend to see what I can find and let you know. Regards, ? ? ? -Paul
|
Thanks Paul. ?Just to let you know, yesterday I swapped out the PC for an older slower one with just pempro and Gemini latest version loaded. Got exactly the same results. ?So that is 4 converters, 3 cables, 2 PCs and 2 control panels. ?Really weird! ?
regards
mark
|
Hi Mark,
Perhaps the error is that PEMPro is sending a command that the earlier control boxes don't understand.
Please use the PEMPro Log zipper on each computer and have it zip the Debug logs for the last 14 days. Attach both zip files (one from each computer) to a post here.
-Ray
toggle quoted message
Show quoted text
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of madastro
Sent: Saturday, March 30, 2024 12:08 AM
To: [email protected]
Subject: Re: [Gemini_users_io] Gemini 1 L4 Issues loading PEC Curve from PemPro 3
Thanks Paul. Just to let you know, yesterday I swapped out the PC for an older slower one with just pempro and
Gemini latest version loaded. Got exactly the same results. So that is 4 converters, 3 cables, 2 PCs and 2 control
panels. Really weird!
regards
mark
|
Hi Ray,?
That would be my first guess is that something is being set or used as if this is a Gemini-2 L6 mount. After a quick check, I don't see anything obvious, though.
Mark -- can you please share a PEMPro .ppc or captured data? .txt file you use for the curve upload? I can try to reproduce the steps here to see if I get the same errors.
Regards,
? ?-Paul
toggle quoted message
Show quoted text
On Sat, Mar 30, 2024 at 12:34 PM, Ray Gralak wrote:
Hi Mark,
Perhaps the error is that PEMPro is sending a command that the earlier control boxes don't understand.
Please use the PEMPro Log zipper on each computer and have it zip the Debug logs for the last 14 days. Attach both zip files (one from each computer) to a post here.
-Ray
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of madastro
Sent: Saturday, March 30, 2024 12:08 AM
To: [email protected]
Subject: Re: [Gemini_users_io] Gemini 1 L4 Issues loading PEC Curve from PemPro 3
Thanks Paul. Just to let you know, yesterday I swapped out the PC for an older slower one with just pempro and
Gemini latest version loaded. Got exactly the same results. So that is 4 converters, 3 cables, 2 PCs and 2 control
panels. Really weird!
regards
mark
|
Ray -- one problem I see is that regardless of what Gemini or the ASCOM driver are configured to for PECMax (usually 6400 steps for Gemini-1/G11), PEMPro tries to retrieve or program 2x the worm cycle, or 12800 steps. Can you please check why? Here's an example, only 6400 steps were programmed in PEC, but PEMPro retrieves this, apparently thinking this is a 2x worm cycle recording:  This is what it should look like:  Regards, ? ?-Paul
|
madastro
Paul Kanevsky