Mesu not tracking on initial testing

[Tom OD]

Ok So I have an update on this issue. I m back from Ollys where I tried to test the RS232 connection.

I brought my Alienware laptop with me as a second Host to test, and for each test the Eagle was supplying power t the Mesu.

So using a  New USB cable:

 

1) I first tried the Alienware connected to the Mesu. After setting the Comm port to 9 as that was what appeared in the Device Manager when I plugged it in, the SiTech Config

program showed a Good Connection and it read the Encoders and other data from the Mount.

I opened the SiTech.exe and it had the "2" errors Faking Servos, and Bad Comms, Then I set the Comm port to 9 in the Misc tab in the Sitech.exe Change Config tab.

On re-opening the Sitech.exe 10 times in a row, there was only the Below Horizon Limit, and Not Initialised seen. All 10 times seemed to work as I think it should.

2) I tried the same with the Eagle and the new USB cable. I checked that the Comm port 7 was Set up in the Config file. However it gave the 2 errors the first time I opened the SiTech.exe program.

 Over to the RS232 / Serial to USB tests then. I had my cable made up, and I was able to confirm that the pins had connectivity with a multimeter. Harry Page and Taj from Sitech confirmed that the pins I had soldered were correct.

 

I opened the cover and set the Jumper to one for RS232 comms. There was a large dead spider in there, so I did wonder if it had fried something when I first saw it.

 

3) I used the Alienware first with the RS232 and it connected through Comm Port 8 seen in the Device Manager, 10 times without issue. Just the Below Horizon and Not Initialised seen.

4) I tried the Eagle and the Comm port was also 8 (I guess that was it holding the port as Odd Socks said). It erred on the first opening in Sitech.exe.

 

I power cycled the Eagle to try again in case it needed that after the Comm port change. It worked 5 times, but failed on the 6th attempt. The second time it has made it that far in all my tests.

So in conclusion as I have not yet Initialised / Set up the mount, or set Cal Init, or Any stars having the Not Intialised and Below Horizon Limit seem normal at this point.

This means that the Alienware can work on both USB and RS232, while the Eagle is unable to work on either setting of comms.

As all tests were done with the Eagle powering the Mesu, power is not the issue given that the Alienware worked each time, every time.

 

A friend has suggested that using the ComDisable feature may hold the port open and allow the comms to work. I have yet to try that though.

If it does not, then I guess I need to remove the Eagle and go to another PC / Laptop source. Hopefully Primaluce can offer some help before any final decisions are made to the set up.

Tom

 

 

[Oddsocks]

Hello Tom.

Away travelling at the moment so access to SGL very limited.....

I don’t know the Eagle other than seeing a few pictures and reading the occasional user posts here on SGL and other forums so this response is not specific to the Eagle, only generic.

From your tests it does sound as though the Eagle is possibly suspect.

Although you have tested with USB and RS232 on the Eagle, both communication methods still rely on and use the Eagles USB ports, the only real difference is that using RS232 forces a slower communication rate, which is generally more robust than straight USB to USB, Eagle to Sitech.

From an engineers stand point the next test I would do, just to eliminate the possibility, would be to check for ground loops and power supply noise using an oscilloscope.

Although your Eagle power supply might have enough amps to power all the kit connected to the Eagle, if the supply is electrically ‘leaky’ or noisy then it’s possible that an earth leakage route can exist that flows from the Eagle’s USB ports, and RS232 adaptor if used, across the connection to the Sitech, and via the Sitechs 0V common down to ground via the mount and pier.

If the power supply is leaking to earth across this route it will cause communication issues.

By using the Alienware instead of the Eagle as the connection to the Sitech you effectively broke any possible power supply leakage route via the Sitech’s com ports.

Appreciating that you probably don’t have access to an oscilloscope for testing for ground loops and noise on the power supply, and that travelling back and forth between home and Ollies is not really an option can you arrange for Ollie to try a different power supply? preferably a clean source for the Eagle such as a big leisure battery, or, if this is not possible and you are using some kind of AC to DC convertor try powering the AC-DC convertor via a site safety isolating transformer as this removes ground leakage route. Ollie may have one already at his site since he has done so much work outside over the years. 

Although a site safety transformer is not suitable for full time usage, they usually only have a ~25% duty cycle and aren’t designed for continuous use or they overheat, but can be used in short periods of twenty minutes or so for testing.

If running on battery, an isolating transformer or a different AC to DC convertor cures the problem then the simplest solution is to prevent possible earth leakage routes via the Eagle-Sitech in the first place, a galvanic (opto) isolator for either the USB - USB or USB to RS232 convertor that is used between Eagle and Sitech would do that but I would hesitate to suggest you thow more money at the problem without proving that a ground loop is to blame first.

Try battery power first for testing, if possible, for the Eagle - Sitech and see how that pans out, if the problem is cured then a galvanic (opto) isolator on the Eagle to Sitech USB-USB or USB-RS232 may be all that is required.

BTW, port # retention on the USB to RS232 adaptor isn’t between computers, only on the same computer, the fact is was the same port # on both the Eagle and the Alienware is just ‘chance’.

HTH.

William.

[steppenwolf]

11 hours ago, Tom OD said:

On re-opening the Sitech.exe 10 times in a row, there was only the Below Horizon Limit, and Not Initialised seen. All 10 times seemed to work as I think it should.

This is all good, Tom and a very positive step forwards. The 'Below Horizon Limit' and 'Not Initialised' messages are not a big problem, they will need to be addressed and I can tell you how to do that when the time comes but at least you now have reliable comms from a PC thus showing that the Mesu is working fine in this regard.

I agree with William that it is the Eagle that is the most cause of the comms issues.

[Tom OD]

3 hours ago, steppenwolf said:

This is all good, Tom and a very positive step forwards. The 'Below Horizon Limit' and 'Not Initialised' messages are not a big problem, they will need to be addressed and I can tell you how to do that when the time comes but at least you now have reliable comms from a PC thus showing that the Mesu is working fine in this regard.

I agree with William that it is the Eagle that is the most cause of the comms issues.

Hi Steve,

It was an anxious few openings and closings of the SiTech before the error arrived, but with the Alienware connecting at least it tells us that the mount is more than likely good, once I go for the full set up that is.

Yes I have your notes on the Below Horizon fix, so I'll be able to clear that Below Horizon I think once I get to that stage.

Tom

[ollypenrice]

Marco and Yves suggested that I try earthing the mount so I've run an earth cable between the mount and the earth pin of a spare plug in the observatory.

I don't have a site transformer here but I do have a spare car battery.

I must say that the reliability of an assortment of astro-dedicated compact computers used here has been lamentable. I would follow Sara, go for a basic desk top with sufficient USB ports and put up with more cables.

Olly

 

[Tom OD]

5 hours ago, Oddsocks said:

If running on battery, an isolating transformer or a different AC to DC convertor cures the problem then the simplest solution is to prevent possible earth leakage routes via the Eagle-Sitech in the first place, a galvanic (opto) isolator for either the USB - USB or USB to RS232 convertor that is used between Eagle and Sitech would do that but I would hesitate to suggest you thow more money at the problem without proving that a ground loop is to blame first.

Try battery power first for testing, if possible, for the Eagle - Sitech and see how that pans out, if the problem is cured then a galvanic (opto) isolator on the Eagle to Sitech USB-USB or USB-RS232 may be all that is required.

Thanks William,

Olly has added an Earth cable from the mount to an Earth pin on a plug, so I'll try later and see if that helps.

We don't have an oscilloscope as far as I know onsite. Is there anyway to try and measure the voltage on the Earth line with a multimeter, would it even give a sensible result?

I power the observatory through a UPS system. So incoming 220v power goes to the UPS. Then the out puts from the UPS feed an extension lead. The Eagle and al other devices are coming from that power strip. Would the UPS not constitute a smooth voltage?

If not I'll get a high Amp battery to test the Eagle. I have only been loading it with the Mesu as the Power load during my testing. All other devices were off.

I need to read up some more on the Galvo Opto option, but I presume I could just solder one in Series onto the USB / RS232 cable. Is only one wire needed to have the opto isolator?

Thanks Tom.

[Tom OD]

20 minutes ago, ollypenrice said:

Marco and Yves suggested that I try earthing the mount so I've run an earth cable between the mount and the earth pin of a spare plug in the observatory.

I don't have a site transformer here but I do have a spare car battery.

I must say that the reliability of an assortment of astro-dedicated compact computers used here has been lamentable. I would follow Sara, go for a basic desk top with sufficient USB ports and put up with more cables.

Olly

 

I was thinking about the car battery too! I used one before as my dew heater power supply

It could at least support a large current needed to fit the Eagle. Of course the caveat to this is that the power input on the Eagle is a CPC type screw in cable. In other words

I would need to hodge-bodge up another cable for you with connectors to fit to the car battery.

I'll update you on that progress.

Tom

[andrew s]

If you have multiple paths to earth this can cause trouble with current flowing in them v to ground. Try and collect all earth points together and connect them all to ground once.

You also want to avoid earth loops where two bits of kit have an earth connection and both are grounded. This often causes hum in audio equipment. 

Regards Andrew 

Edited March 28, 2019 by andrew s

[Oddsocks]

1 hour ago, Tom OD said:

Is there anyway to try and measure the voltage on the Earth line with a multimeter, would it even give a sensible result?

 

1 hour ago, Tom OD said:

The Eagle and al other devices are coming from that power strip. Would the UPS not constitute a smooth voltage?

Hello Tom.

Most likely a multimeter wouldn't help, leakage currents are typically too small to measure sensibly, an oscilloscope is the best method, and that is what I would have used for this kind of problem when I used to work on these sorts of problems for a living.

The UPS provides smooth 220/230 AC to the Eagle, and I have an APC UPS in my own observatory to protect the equipment and shut down the computer correctly in the case of a power interruption, the problem then is what the Eagle does with that clean supply. The step-down power convertor, AC/DC, that is used in general electronics, switch mode PSU's, can leak a small amount of incoming mains 220/230VAC to the 0V rail of the DC output via internal interference suppression capacitors. The current is very small, just a few milliamps, but if this current finds a route to earth via USB or RS232 leads it 'can' interfere with comms. If you have an iPad, or a similar device with a metallic case you will often feel a tingle when running your fingers over the device when it is connected to an AC/DC charger, this is quite safe, and is caused by the leakage current from the AC/DC convertors suppression capacitors as described above, but it 'can' be the source of comms issues when connecting lots of different devices together.

I think the MESU/Sitech combination maybe prone to these sorts of issues because the Sitech control unit is solidly bolted to the MESU and the MESU is bolted to the pier, which is grounded, so it is the last link in the chain for a leakage current to flow to earth.

If leakage currents are the issue then there are several possible approaches to the problem. First, all separate pieces of equipment can be connected to a single earth point, but this means the MESU has to be installed on the pier using isolator washers, or an insulating sheet between the MESU and the pier, with isolator washers also used under the bolt heads that bolt the MESU to the pier and isolation ferrules or plastic tape around the parts of the bolts that pass through the MESU that bolt the MESU to the pier. If this is done then there can be no leakage currents flowing between the Eagle - Sitech - Pier - Earth, so long as the MESU and pier remain dry, in an observatory this is easy to control but out in the open, if the MESU and pier become damp through frost and condensation etc then the isolation may break down leading to intermittent comms issues that are difficult to trace. With the Sitech/MESU isolated from ground and all devices connected to a single earth point then earth loops and leakage currents can be controlled. The second way is to use galvanic isolators between the different components of the system.

As a point of reference, I have a Paramount MX and it has had two failed MKS5000 control boards in five years, at $800 a time this is not funny. The problem is the USB UART on the MKS5000 is totally unprotected against voltage spikes, if an active thunderstorm passes close by the observatory, the induced voltage spikes generated in the long USB leads by the lightning destroys the USB UART. After the second MKS5000 failure I fitted a USB opto-isolator, that was three years ago and I haven't had any further failures since then, I also bought a few spare USB UART chips, just in case, as the cost of buying the replacement boards from Software Bisque was not justifiable, especially since they could have chosen to fit rapid spike suppression diodes on the USB inputs that would have given a degree of protection.

The type of isolator I have on the MX is one of these:

https://uk.farnell.com/olimex/usb-iso/usb-isolator-1000vdc-for-pc-laptop/dp/1795095?st=usb isolator

It gets it's power from the host USB socket (Eagle), via a built-in isolation supply, but if this proves insufficient you can plug in a +8V to +15V external supply. This isolator is connected via a very short USB lead ~15cm long, to the USB input of the Paramount with a long 3mtr USB cable from the isolator back to the host computer.

These isolators are built for use in the dry, so if the Sitech/MESU gets to be damp then put it in a box.

As mentioned before, I wouldn't jump to a conclusion that a leakage current is to blame and rush out to buy one without proving that this is the problem first and powering from a battery would be the simplest way.

I will be away from the internet now till the weekend, probably Sunday, or early next week, but I'll look back in when I get a chance.

William.

[ollypenrice]

2 hours ago, Oddsocks said:

 

Hello Tom.

Most likely a multimeter wouldn't help, leakage currents are typically too small to measure sensibly, an oscilloscope is the best method, and that is what I would have used for this kind of problem when I used to work on these sorts of problems for a living.

The UPS provides smooth 220/230 AC to the Eagle, and I have an APC UPS in my own observatory to protect the equipment and shut down the computer correctly in the case of a power interruption, the problem then is what the Eagle does with that clean supply. The step-down power convertor, AC/DC, that is used in general electronics, switch mode PSU's, can leak a small amount of incoming mains 220/230VAC to the 0V rail of the DC output via internal interference suppression capacitors. The current is very small, just a few milliamps, but if this current finds a route to earth via USB or RS232 leads it 'can' interfere with comms. If you have an iPad, or a similar device with a metallic case you will often feel a tingle when running your fingers over the device when it is connected to an AC/DC charger, this is quite safe, and is caused by the leakage current from the AC/DC convertors suppression capacitors as described above, but it 'can' be the source of comms issues when connecting lots of different devices together.

I think the MESU/Sitech combination maybe prone to these sorts of issues because the Sitech control unit is solidly bolted to the MESU and the MESU is bolted to the pier, which is grounded, so it is the last link in the chain for a leakage current to flow to earth.

If leakage currents are the issue then there are several possible approaches to the problem. First, all separate pieces of equipment can be connected to a single earth point, but this means the MESU has to be installed on the pier using isolator washers, or an insulating sheet between the MESU and the pier, with isolator washers also used under the bolt heads that bolt the MESU to the pier and isolation ferrules or plastic tape around the parts of the bolts that pass through the MESU that bolt the MESU to the pier. If this is done then there can be no leakage currents flowing between the Eagle - Sitech - Pier - Earth, so long as the MESU and pier remain dry, in an observatory this is easy to control but out in the open, if the MESU and pier become damp through frost and condensation etc then the isolation may break down leading to intermittent comms issues that are difficult to trace. With the Sitech/MESU isolated from ground and all devices connected to a single earth point then earth loops and leakage currents can be controlled. The second way is to use galvanic isolators between the different components of the system.

As a point of reference, I have a Paramount MX and it has had two failed MKS5000 control boards in five years, at $800 a time this is not funny. The problem is the USB UART on the MKS5000 is totally unprotected against voltage spikes, if an active thunderstorm passes close by the observatory, the induced voltage spikes generated in the long USB leads by the lightning destroys the USB UART. After the second MKS5000 failure I fitted a USB opto-isolator, that was three years ago and I haven't had any further failures since then, I also bought a few spare USB UART chips, just in case, as the cost of buying the replacement boards from Software Bisque was not justifiable, especially since they could have chosen to fit rapid spike suppression diodes on the USB inputs that would have given a degree of protection.

The type of isolator I have on the MX is one of these:

https://uk.farnell.com/olimex/usb-iso/usb-isolator-1000vdc-for-pc-laptop/dp/1795095?st=usb isolator

It gets it's power from the host USB socket (Eagle), via a built-in isolation supply, but if this proves insufficient you can plug in a +8V to +15V external supply. This isolator is connected via a very short USB lead ~15cm long, to the USB input of the Paramount with a long 3mtr USB cable from the isolator back to the host computer.

These isolators are built for use in the dry, so if the Sitech/MESU gets to be damp then put it in a box.

As mentioned before, I wouldn't jump to a conclusion that a leakage current is to blame and rush out to buy one without proving that this is the problem first and powering from a battery would be the simplest way.

I will be away from the internet now till the weekend, probably Sunday, or early next week, but I'll look back in when I get a chance.

William.

We are most certainly in a thunderstorm zone in the summer months. William in Dorset will not be anywhere near our level of risk, I don't think. Our thunderstorms are big ones.

Olly

[Tom OD]

3 hours ago, Oddsocks said:

 

Hello Tom.

Most likely a multimeter wouldn't help, leakage currents are typically too small to measure sensibly, an oscilloscope is the best method, and that is what I would have used for this kind of problem when I used to work on these sorts of problems for a living.

The UPS provides smooth 220/230 AC to the Eagle, and I have an APC UPS in my own observatory to protect the equipment and shut down the computer correctly in the case of a power interruption, the problem then is what the Eagle does with that clean supply. The step-down power convertor, AC/DC, that is used in general electronics, switch mode PSU's, can leak a small amount of incoming mains 220/230VAC to the 0V rail of the DC output via internal interference suppression capacitors. The current is very small, just a few milliamps, but if this current finds a route to earth via USB or RS232 leads it 'can' interfere with comms. If you have an iPad, or a similar device with a metallic case you will often feel a tingle when running your fingers over the device when it is connected to an AC/DC charger, this is quite safe, and is caused by the leakage current from the AC/DC convertors suppression capacitors as described above, but it 'can' be the source of comms issues when connecting lots of different devices together.

I think the MESU/Sitech combination maybe prone to these sorts of issues because the Sitech control unit is solidly bolted to the MESU and the MESU is bolted to the pier, which is grounded, so it is the last link in the chain for a leakage current to flow to earth.

If leakage currents are the issue then there are several possible approaches to the problem. First, all separate pieces of equipment can be connected to a single earth point, but this means the MESU has to be installed on the pier using isolator washers, or an insulating sheet between the MESU and the pier, with isolator washers also used under the bolt heads that bolt the MESU to the pier and isolation ferrules or plastic tape around the parts of the bolts that pass through the MESU that bolt the MESU to the pier. If this is done then there can be no leakage currents flowing between the Eagle - Sitech - Pier - Earth, so long as the MESU and pier remain dry, in an observatory this is easy to control but out in the open, if the MESU and pier become damp through frost and condensation etc then the isolation may break down leading to intermittent comms issues that are difficult to trace. With the Sitech/MESU isolated from ground and all devices connected to a single earth point then earth loops and leakage currents can be controlled. The second way is to use galvanic isolators between the different components of the system.

As a point of reference, I have a Paramount MX and it has had two failed MKS5000 control boards in five years, at $800 a time this is not funny. The problem is the USB UART on the MKS5000 is totally unprotected against voltage spikes, if an active thunderstorm passes close by the observatory, the induced voltage spikes generated in the long USB leads by the lightning destroys the USB UART. After the second MKS5000 failure I fitted a USB opto-isolator, that was three years ago and I haven't had any further failures since then, I also bought a few spare USB UART chips, just in case, as the cost of buying the replacement boards from Software Bisque was not justifiable, especially since they could have chosen to fit rapid spike suppression diodes on the USB inputs that would have given a degree of protection.

The type of isolator I have on the MX is one of these:

https://uk.farnell.com/olimex/usb-iso/usb-isolator-1000vdc-for-pc-laptop/dp/1795095?st=usb isolator

It gets it's power from the host USB socket (Eagle), via a built-in isolation supply, but if this proves insufficient you can plug in a +8V to +15V external supply. This isolator is connected via a very short USB lead ~15cm long, to the USB input of the Paramount with a long 3mtr USB cable from the isolator back to the host computer.

These isolators are built for use in the dry, so if the Sitech/MESU gets to be damp then put it in a box.

As mentioned before, I wouldn't jump to a conclusion that a leakage current is to blame and rush out to buy one without proving that this is the problem first and powering from a battery would be the simplest way.

I will be away from the internet now till the weekend, probably Sunday, or early next week, but I'll look back in when I get a chance.

William.

Thanks William,

That is an excellent summation. Post any battery testing we do,

I think insulating the mount / Pier would require a full rebuild of everything. I m not sure that is an option, given the way we have got the Mesu onto the pier and bolted it all down.

I would go with the second option, and the opto isolator.

After Olly connected a ground from the Earth pin on the plug extension to the Mesu, I tried again, and it failed again. If I have understood you correctly, this would not work anyway, as we would be givng the stray voltage a path to ground this way, rather than isolating it from ground as in option 1 as you described.

Time for me to ask Olly to charge the batteries

Tom

 

 

[Tom OD]

Well the latest update is very odd.

I logged in today and the Eagle was showing a battery % and remaining time etc in the Ui. The eagle manager was the same version as before, I thought it might have updated the rev and added the graphic of the power %.

I had 2 other astronomers onsite last week who I asked to remove the USB/SERIAL/RS232 cable and to remove the RS232 jumper on the board, and reinstall my new USB cable (which I had tested as working with the Alienware but not with the Eagle).

Anyway now the Eagle connected 10x with the battery running. I figured maybe the UPS had kicked in during a power cut or disconnection, and again later after the AC came back on it connected 10x without an issue.

The Earth cable connection Olly added a few days ago, made no difference when I tested it.

I cant understand why it works, but I d still like to know, as I m nervous it could return again. I fear we ve pulled a "Homer" here, a success through idiocy. Maybe I should not complain, but I remain cautious. I ve sent a few questions to

the guys involved in the replacement of the cables just in case something was routed differently etc..

I'll still send over the opto isolator I think for installation just in case.

More to come on all this I think.

Tom.

 

[Oddsocks]

Has all the hall marks of an earth leakage / ground loop, Tom.

Just a small nick in the jacket insulation of your old USB cable to the Sitech can introduce an unwanted ground point, conversely, a broken shield on the USB cable would introduce noise on the USB signal lines.

When you tested the RS232 adaptor with the Eagle were you using the original USB cable as a connection beween the Eagle and the RS232 adaptor or bypassing the old cable completely? Leaving the old USB cable in situ or disconnecting both ends while testing the RS232 adaptor etc.....?

I meant to ask earlier whether your UPS was an on-line / line interactive or off-line / standby model. I have had problems with several makes of the on-line / line interactive types over the years where the output AC is not symmetric around earth and this can lead to small differences between the DC 0V levels of your connected AC/DC step down convertors and true ground.

We used opto or magneto galvanic isolators quite extensively for both USB and serial interconnections on medical equipment to prevent earth loops, though more to do with connecting between equipment on different mains phases in three phase environments we also found them necessary on systems using a UPS (the isolators we used, being medical rated, cost several hundred Euros each).

Having the UPS run through an operating cycle, mainly the off-line types, is something we did every three months on maintenance inspections as this is important to help keep the UPS’s internal change-over contactor contact-set clean, if the UPS was left for many months always in the standby mode then the change-over contactor became electrically noisy due to contact tarnishing which introduces high frequency noise and spikes on the output side.

While you wait for the USB isolator, if the problem occurs again on mains supply ask Ollie to temporarily take the UPS out of circuit, if the UPS uses standard IEC leads just shut it down, pull the input and output leads and plug them together. It would be another possible problem source eliminated.

William.

[Tom OD]

2 hours ago, Oddsocks said:

Has all the hall marks of an earth leakage / ground loop, Tom.

Just a small nick in the jacket insulation of your old USB cable to the Sitech can introduce an unwanted ground point, conversely, a broken shield on the USB cable would introduce noise on the USB signal lines.

When you tested the RS232 adaptor with the Eagle were you using the original USB cable as a connection beween the Eagle and the RS232 adaptor or bypassing the old cable completely? Leaving the old USB cable in situ or disconnecting both ends while testing the RS232 adaptor etc.....?

I meant to ask earlier whether your UPS was an on-line / line interactive or off-line / standby model. I have had problems with several makes of the on-line / line interactive types over the years where the output AC is not symmetric around earth and this can lead to small differences between the DC 0V levels of your connected AC/DC step down convertors and true ground.

We used opto or magneto galvanic isolators quite extensively for both USB and serial interconnections on medical equipment to prevent earth loops, though more to do with connecting between equipment on different mains phases in three phase environments we also found them necessary on systems using a UPS (the isolators we used, being medical rated, cost several hundred Euros each).

Having the UPS run through an operating cycle, mainly the off-line types, is something we did every three months on maintenance inspections as this is important to help keep the UPS’s internal change-over contactor contact-set clean, if the UPS was left for many months always in the standby mode then the change-over contactor became electrically noisy due to contact tarnishing which introduces high frequency noise and spikes on the output side.

While you wait for the USB isolator, if the problem occurs again on mains supply ask Ollie to temporarily take the UPS out of circuit, if the UPS uses standard IEC leads just shut it down, pull the input and output leads and plug them together. It would be another possible problem source eliminated.

William.

Thanks yet again William.

When I tested the RS232 cable I removed the old usb cable, and installed just the Star tech serial to USB cable along with my soldered cable. This failed in the Eagle but worked in the Alienware.

I also tested both Eagle and Alienware with a brand new fancy looking USB cable. Again Eagle failed but Alienware worked.

 

I m not sure which model UPS I have. I'll get Olly to check the manual which I left in the Observatory.

Olly has said there was no power failure onsite yesterday that he was aware off. Also I don't think I could have used Team Viewer to access my laptop as I m connected to the main observatories Network. While this is on a UPS too, the house feed which we all use is not on a UPS, so any power failure should have killed the net period.

I think my UPS just kicked in, not sure why. Better be lucky than good right :)

I think I'll take your advice and just run the UPS every now and again to cycle the system in case of a noise build up.

The Opto Isolator will be ordered too. As soon as Farrell take card payments. Their website is not accepting my CC numbers.

Tom

[Oddsocks]

2 hours ago, Tom OD said:

Olly has said there was no power failure onsite yesterday that he was aware off.

Probably best to ask Ollie to give the IEC power connection cables to the UPS a once over, they never seem to fit tightly in the sockets and can be easily dislodged, it could be the mains input cable has worked loose, equally, it could just be a temporary reduction in mains input voltage that triggered the UPS into action, depending which model UPS you have and what thresholds have been set up a reduction of just ten Volts or so below nominal mains input may be enough to trigger the UPS while not being noticeable to the residents on site.

Farnell do have the occasional glitch with their payments system and I find it best to give their order line a call when having problems though normally once your account is set up and the CC registered then ordering is quite straight forward.

I'll keep watching the thread to see how things pan out.

William.

[Tom OD]

On 29/12/2018 at 15:19, PhotoGav said:

Open SGP and connect to camera and Mesu

Open up CDC and connect to Mesu

Click ‘Start’ in SiTech exe (rather than Unpark), that sets it into Blind Tracking mode

Move the Mesu using the buttons in SiTech exe to a convenient part of the sky, not too near the NCP

Do a Blind Solve from within SGP

When it solves the image, SiTech will bring up its Init dialog box, Click on Offset Init

The mount should now be initialised and tracking properly

Slew the Mesu to your chosen park position using the buttons in SiTech exe

Click Stop in SiTech exe

Click Set Park Position

Click Park

So I now have the Opto Isolator in series with the USB to the Eagle, it connects without errors. I m too nervous to try it ten times in a row, but its working each time I connect so far.

So now I moved onto the next step (Gavs guide above), getting the Offset Init working and the mount tracking and slewing. I'll do the Cal Init later with a proper map of the sky.

Last night I was just starting the CDC and seeing if I could blind solve. Despite the bright moon it could solve on Lum filter, but only if i took a 5s pic.

CDC showed the cross hair on the map, and I just did a Goto M13 without any Offset Init. Again I was just messing about. Anyway the mount slewed for the first time and M13 was in the frame almost centered.

So this was a big step. However when it was saying tracking in CDC Connect, it was alternating between Slewing / Tracking on the Sitech.exe program. See first pic.

 

I tried again tonight, this time following the steps above, but it would not solve, so no Offset Init again.

The moon was too bright and I was not Blind tracking on Start. Again it was Slewing / Tracking. See screenshots, so I had to hit "stop".

I probably need to wait for the moon to go away to take a long enough exp without trailing stars, but any idea why its just not tracking when I hit start. Clearly the squiggly pic shows the mount is moving / making corrections when

it should be at least Blind tracking right?

Thanks Tom.

 

[PhotoGav]

Mmm, strange... why is it slewing? Where have you told it to slew? What happens when you just start SiTech and then click Start, with no CdC involved?

[harry page]

Hi

I have a mesu and there is a bug in the sitech controller , in which when I first connect it decides to slew off wildly for no reason

does drive me potty and only does this once in a while , the sitec controller is good but also is buggy

sometimes after stopping tracking , I can not get it to start again without a reboot of the controller !!!!!

I must try and find out why it does this but have found its difficult to get these bugs resolved

anyway enough off my winge , anybody else have these

Harry

[tomato]

My Sitech can very occasionally refuse to slew the mount when requested, requiring a restart. As I stay with the kit during a session it’s annoying  but not a catastrophe, it would obviously be a major deal if I was properly remote.  I also set up and take down every time so some of my USB plugs and sockets are getting worn, it wouldn’t surprise me if that was behind it.

[Tomatobro]

Having spent years working with and designing instruments that use micro processors the one thing I have learnt is that most start up issues are the result of how the power is applied. This falls into two areas. One is the power supply and the second is how the power is applied.

If the power is applied by pushing in a socket there is a good chance that the power will apply and disconnect many times during this action and can impact on the start up of the circuit. If at all possible connect everything up first then apply power via a good quality switch.

On the power supply having some headroom on the current output vs. requirement is a help as charging capacitors etc place an initial load during connection.

Its true that modern processors have inbuilt start up procedures to allow for stabilisation but they can be defeated. Probably that is why the problems only occur now and again.

microprocessors live in a different world to the rest of us mere mortals

[Tom OD]

On 21/04/2019 at 17:22, PhotoGav said:

Mmm, strange... why is it slewing? Where have you told it to slew? What happens when you just start SiTech and then click Start, with no CdC involved?

Without CDC connected, it just says Blind Tracking.

Seems that as soon as I connect CDC, that on Start or Connect in CDC, then it starts this Slewing / Tracking. I m waiting for clear skies to test this further.

Tom

[PhotoGav]

2 hours ago, Tom OD said:

Without CDC connected, it just says Blind Tracking.

Seems that as soon as I connect CDC, that on Start or Connect in CDC, then it starts this Slewing / Tracking. I m waiting for clear skies to test this further.

Tom

Good, it’s doing the right thing on its own, so it must be an option or setting or something unusual from CDC. I’m sure this will be solvable when clear skies return.

[Tom OD]

23 hours ago, Tomatobro said:

Having spent years working with and designing instruments that use micro processors the one thing I have learnt is that most start up issues are the result of how the power is applied. This falls into two areas. One is the power supply and the second is how the power is applied.

If the power is applied by pushing in a socket there is a good chance that the power will apply and disconnect many times during this action and can impact on the start up of the circuit. If at all possible connect everything up first then apply power via a good quality switch.

On the power supply having some headroom on the current output vs. requirement is a help as charging capacitors etc place an initial load during connection.

Its true that modern processors have inbuilt start up procedures to allow for stabilisation but they can be defeated. Probably that is why the problems only occur now and again.

microprocessors live in a different world to the rest of us mere mortals

Adding in the opto isolator seems to have too so far. I have a battery power supply ready to go in case further testing is needed.

Tom.

[Tom OD]

At this point I could almost ignore CDC, and just blind track and use the Sitech buttons to send the mount to the RA and Dec that I wanted, once I could do a Blind Solve.

At least I know the Blind Solve works too, or did last week before I connected the CDC to the mount.

T.

[Tom OD]

Another update on this but its just pure frustration. I cannot get the mount to work with CDC.

If I dont connect to CDC and just press Start, I get blind tracking. After about 3 attempts it will blind solve, but once I connect to CDC, then it starts the Slewing / Tracking alternation when I start to get the squiggly lines in the pic attached.

Even after disconnecting the camera and mount and Sitech it still gave me Slewing / Tracking when i hit start. I had to restart the Eagle just to get back control. Nothing changed. It looks like I can not connect CDC and only get Blind tracking, or connect CDC and have it completely messed up with the slewing / tracking. I have no idea what the hell is going on here. This is really starting to [removed word] me off now. I ve gone from one  ridiculous electrical / software issue to another ridiculous software / mechanical issue. 

If anyone can offer any more help I d be grateful, go ahead and dial in to the system and have a go yourself if you d like, as right now, and for the last 4 months I ve been staring at a shed that has just been sitting there doing nothing when it has so much potential.

As it turns out today is the 40th Anniversary of Alien. My trackign seems to stencil out a pretty good Alien head of whatever star it thought it was looking at.

Tom.

Tom