Hi folks me again. There’s a problem that’s been occurring a fair bit lately. For some reason, the mount (NEQ6) disconnects from SGP (the connect button turns grey from orange) during imaging sessions for no apparent reason. However EQMOD is still connected and PHD2 is still guiding. So when it comes to slew, meridian flip or park, SGP therefore doesn’t do any of that and neither does it reconnect to the mount, which is strange because I can manually click the connect button and the mount reconnects to SGP. I’ve only noticed the problem since moving from using a hand controller in “PC direct mode” to a direct connection with a shoestring cable.
Has anyone else seen this problem or have an idea as to what is going on. The problem becomes a pain for unattended sessions.
I’m running the latest beta version (4.0.0.661) of SGP.
Here’s a dropbox link to the logs.
Looks like the disconnection happened at time 23:21:29 according to the logs. When I was in bed!!
Thanks for any input.
I have had issues like this, SGP disconnected but EQMOD and PHD2 seemingly happy. The first thing to check is the reliability of the 12V supply, my scope had the unreliable DC-jack so I replaced it with a screw locking connector; the second thing I did was to disconnect any other software like Cartes du Ciel while imaging to reduce any potential impact of multiple devices polling.
I haven’t had any problems since these changes.
Ok good to know!! Thanks for the input. Usually it’s just SGP and PHD2 running but occasionally I have stellarium going so I’ll make sure that’s always shutdown before I leave it for the night. When you say “reliability of the 12V supply” are you referring to voltage drop, loose connections, etc. When imaging at home, I use a 12V 7.5A supply from mains power. The NEQ6 has a 2.1mm dc socket but not a screw locking connector so I’m guessing that would be a modification job.
I’m guessing that replacing the dc jack with a screw connector involves a bit of electrical know how. Or at least knowing how to solder?
I also use a strong mains supply so I didn’t suspect that. I was able to wobble the dc jack and observe the light flicker so I thought that mod worth doing. Yes it does need solder skills and some bravery 
here’s a starter link
Alternatively you could try securing the jack with a velcro strap if it’s a bit loose
I don’t think that was the main cause though, I have worked on a scope driver before so I knew that it’s quite possible to miss commands if there’s too much going on, that’s why I tried disconnecting CdC once a sequence had started.
I’d recommend trying this before the power connector mod.
A further point to consider is the voltage drop along the cable run between your PSU and equipment. It is easy on the web to find tables for resistance per metre for various guages of copper wire. I suspect shop-bought power supply cables are typically 16 or perhaps 14 AWG. The resistance per metre of such cables is circa 0.02 ohms per metre. I like to use a 10 metre cable from my indoors PSU (rated 6-8 amps @ 13.8V) to my mount etc in the garden via a Pegasus Power box, so resistance of about 0.2 ohms without including resistance of plug-in connectors etc. So I see the voltage recorded on the Pegasus drop from 13.7 V to 13.0 V when drawing about 3 amp, i.e. From Ohms Law: V= I x R. Voltage drop on the cable run = 0.2 ohm x 3 amp = 0.6 volts
If I try to power a NUC direct from the Pegasus, which adds 2.5 - 3 Amp extra current draw, the measured voltage then falls to c 12.3 - 12.4 V which is insufficient. I am looking to get some 10 or 12 AWG cable sold for loudspeaker connections which I calculate should reduce the voltage loss on the cable by about 50%, result happiness I hope.
Moral: Think about the power draw of your remote equipment and keep PSU cable runs as short as possible and/or obtain suitable guage cables.
1 Like
Thanks Laurie! I have a mate who did a hypertune and belt mod on my EQ6 who said he can also do the power connector mod so I’ll get him to do that. For last nights session, I secured the dc jack with velcro (funny you suggest that now!) and didn’t have a problem all night. On examination, the dc jack on it’s own isn’t tight and easily wobbles around so I can how that might cause the power to bounce around and affect connectivity. Thanks for the suggestion to secure it. We’ll see how we go and if the problem doesn’t return after a few nights session then I can be more certain the dc jack was the problem.
Thanks for the reply. Much appreciated. The PSU is with the telescope and cable length is minimal (less than 1 meter). There’s a voltage meter just before the main ciggie socket that powers the mount and it reads around 12.8V so the mount should be happy with that!
It’s always interesting to hear how 12V devices actually don’t like 12.0V but prefer voltage closer to 13V or over. I’m got friends who’s hardware all say “12V” but in fact some like about 12V but others actually need about 13.5V. Just means they need regulators for the devices.
Your calculations for voltage drop in the power cord are almost correct. Everything you did is perfect except the length of the power cable. A 10 metre cable is actually 20 metres of wire because you have two wires in the cable. This doubles the resistance of the cable, doubling the voltage loss between the power supply and the mount. 20m of 18ga wire has .42 ohms resistance, 16ga has .27 ohms, 14ga has .17 ohms, and 12ga has .11 ohms. These resistance numbers were rounded to two decimal places and taken from an online calculator using 1m = 3.3 feet.
Hi,
You are absolutely right. But in a way this emphasises my point regarding the potential for significant voltage drop on lengthy cable runs
It is a difficult point to prove but I am suspicious that quite a number of users’ reports of intermittent equipment and/or USB failures may well be due to power supply voltage at their equipment being at or below the required threshold level.
I think people should at least be aware of the problem, especially if using typical commercially available 12V extension cables.
Regards
Mike