I tried installing APT however it installed normally but did not resolve the SGP issue - no idea why my installation is different! I've now un-installed it and had a look at windows restore however it doesn't show any APT files as remaining after the uninstall.
This is all very strange, windows environments are a bit of a minefield. I have created a debug version of my driver and passed this to Jared at Main Sequence Software. This debug driver looks identical to the normal driver from the client point of view over the ASCOM API but has all the hardware specific USB communications stubbed out. This allows Jared to install and test the driver without any hardware connected.
I agree Dave. Seems each pc is a bit unique. I'm going to give it a try on my Win7 32bit machine tomorrow but I've realised I can always use your client to rack the focuser in/out if necessary. I'm pretty sure the SGP guys will figure out what's happening with this. They have been incredibly responsive to everyones issues and requests so far.
I will be interesting to see if the issue happens with your debug driver - does the motor delay get reproduced?
Yes it will be interesting. I don't really mind where the issue is as long as it gets found and fixed. If it is in the driver I'll probably end up learning something new which is always good.
No the motor delay is not reproduced at the moment. I am now thinking it's not about the delay. When I was reproducing the fault it was the 100 to 101 boundary that caused the issue ... very odd. The SGP guys did mention they have a bug, something to do with the step size being 100 seem like too much of a coincidence. I thought maybe my driver was incorrectly reporting the step size and generating this bug they mentioned but I can't find any evidence of that. I'm sure Jared will spot it and we can tie it off.
As a matter of interest do you also find that 100->101 causes the problem?
I've found it occurs at bigger steps (between 600-800) however it isn't consistent. Never an issue below 100 unless I press the move button too fast/too many times.
I'm sure you picked this up from the SGP (Mainstream SW) forum. I have fixed your problem with SharpSky, SGP & W7 64bit - it was fun. Went to my parents as Mum has a 64bit machine, plugged everything in and sure enough there was your reported SGP problem ! Installed all the C# developer tools and got to work. I am still not sure if the problem is not six of one and half a dozen of the other. Basically SGP is doing something I was not expecting and have not seen before with other clients and my driver wasn't happy with it. Why it only happens on a W7 64bit machine I have no idea so am hoping the Mainstream guys might know more.
Upshot is I have made the driver more robust in the offending area so with a new driver your problem should be history. If you could afford me a little more time that would be great. I just want to do a bit more testing just to make sure nothing nasty has crept in.
I found your message earlier in this thread on how to uninstall the driver and now have the new one installed. Tested it with SGP and all is working perfectly now. Thanks!
Peter
Last edited by peter_4059; 24-09-2012 at 05:50 PM.
Received my kit earlier this week from Dave, took just over a week to get here. Easy to construct but I used a different motor to what Dave supplied and I required some help to get it running properly. Dave provided prompt advice and after a couple of emails all is running perfectly.
Great kit and software, excellent service, highly recomended.
On the motor front any 4 phase unipolar motor should work fine with the SharpSky controller including the MoonLite Hurst motor. Even though motors have similar specifications due to variations in mechanical construction their response to being driven can be somewhat different. In order to support a wide range of motors I included a motor drive period parameter on the SharpSky properties window. This defaults to 35ms which gives good torque and speed for the motor I have been using. For Neale's motor this had to be increased to 60ms for smooth operation.
I usually increase the value to about 80ms and then reduced until the motor just starts missing steps or chattering and then add 10%.
Today I got around to put together SharpSky kit I received from Dave few days earlier. Kit now include larger stepper motor and pre-made Mini DIN leads. Real bonus for me as my hands is not steady as it used to be and I don’t see that well. In no time PCB was finished although I did more guess then see, where I solder. But then I got over 45 years experience in repairing and manufacturing electronics. On power up everything worked as it should. I decided not to use LEDs except the heart beat one. My scope and mount already resembles a Christmas tree and I did not want to add more flashing lights. All what remains to finish it off is to manufacture focuser bracket. But it is very hot in my tin garden shed I use for mini workshop. Maybe tomorrow. The kit is one of the best I have ever seen and recommended to anyone who wants remote focuser control.
Fantastic news, a new toy just in time for Christmas
Very pleased you liked the kit - half the fun is in the building ! Be very interested to know how you get alone using the design in anger and any suggestions very welcome.
hey dave, seeing some issues with ascom which i think is to do with the sharpsky, both with the sharpsky client and SGP. most nights at some random point, the focuer stops responding to any communications and requires ascom to be complelly shut down to come back to life. this ive seen happen even when not sending commands.
Apologies for the delay, been away at the in-laws.
Mmmm, that's not great news. I have just finished a 24hr soak test with MaxIm DL and have just started a similar test with SGP. I have had issues with SGP pro in the past but thought everything was now running smoothly. SGP uses lots of multi-threading and it's a little difficult to see exactly what is going on sometimes.
I will complete a soak test and see what I can find. I assume you are using the latest version of the driver V3.8 ?
Thanks for you patience, I have been looking into your issue with SGP. I am confident the problem is now resolved. This is what happened .... over the last couple of driver versions I have been making behind the scenes change to make the driver 100% ASCOM 6 and and clean as I can. However, I think that one such change has introduced a problem which I intend to follow up with the ASCOM guys. Upshot is I hope to have fixed the instability and have uploaded a new V3.10 version of the driver to the link on the SharpSky Kit website. I have soaked tested this version over a 24hr period with SGP & MaxIm DL.
The new driver has a new feature under the 'hardware control' panel that allows the user to pre-set the focuser position to any arbitrary value. This does not move the motor but simply changes the value reported in the client to a value you want to see or maybe a zero point. Also the maximum range of the focuser has been extended from 50,000 to 150,000 steps.
Please let me know if the new driver version fixes your issue.
After closing all ASCOM applications uninstall the previous driver before installing the new version. You should see version 3.10 in the SharpSky setup window prior to connection.
So I think there might be a firmware bug or another driver bug, ive currently got ascom connected to the sharpsky kit, telling it to dance around until for some reason it now says when in sharpsky setup via the ascom chooser, PIC firmware version 0.0 and it wont do any movements. However it does not cause an ascom driver crash (yet at least.)
reconnecting usb clears it up, but it does not throw any errors prior to that that are visible
Last edited by technofetishism; 30-12-2012 at 11:25 AM.
Once installed the V3.10 setup screen will look as attached screen shot with default values for MaxStep and Max focuser position at 150,000. Also you should see the new field 'Initial focuser position' with default value of 25000 for a fresh driver install. PC driver version should = 3.10 as screen shot. The reason Max Increment is at 50,000 is this is the largest single step jump you can make in one go.
If you don't get the above it's possible the old driver did not uninstall due to a client dependency, ie a client was open when the old driver was uninstalled preventing Windows from uninstalling cleanly - this happens a lot and is a bit of a pain. Check all clients are closed before uninstalling the old driver.
Can the max increment of focuser also be able to be set to 150,000. Using zero focuser from SGP stops after 50,000 steps but still sets the focuser position to 0.
Currently the max increment is limited to 50,000. So you can move by 50,000 steps in one single movement. This value is passed to the client so SGP should not allow you to move by more than this limit.
Do you want to move by more than 50,000 in one go ?
