Jump to content
  • 1

Loosing USB connection all the time


lukepower

Question

Hello,

 

I noticed today how sensitive my mount/scope combination is regarding balancing. It worked quite well for a few hours, but then there was simply no way to use it. I tried to rebalance it, and it was damn close to perfect, but still I kept loosing the connection. Please note that the tracking (or slewing) continued even without USB connection, so it wasn't a power failure, and I could guess that the balance is not the issue here.

It could be a USB cable issue, even if I use right now a shielded one with an industrial USB hub in front of the scope (and all other accessories working well even during the interruption, like the CCD camera). Maybe it has something to do with the interface to the mount (which I hope is not the case, as I am not really willing to unmount everything to send in the mount :P ).

 

As for the errors, sometimes i get ReadInt errors, sometimes I loose one axis, but most often Autoslew starts to do the Software USB Unplug/Plug procedure, mostly without success.

 

Any ideas?

 

Thanks

Link to comment
Share on other sites

  • Answers 93
  • Created
  • Last Reply

Top Posters For This Question

Top Posters For This Question

Posted Images

Recommended Posts

  • 0

Could be a conflict. Particularly if you are using virtual COM ports over USB for other stuff. What other USB equipment is connected? Try disconnecting one at a time and see if the problem goes away. I had this issue when running a virtual COM port over a virtual USB port over WiFi (possibly unsurprisingly!). When I disconnected that item all my random USB problems went away

Link to comment
Share on other sites

  • 0

Hi Nigel,

 

thank you for your idea. Well, I only have a AAG_CloudWatcher on an USB-to-COM converter (attached to the PC).

On the Hub in front of the mount there is only the GPS stick, and on the mount the ASA stuff (well, the mount itself, the mirror covers), a Starlight focuser (which is straight to USB) and the CCD camera. I will try to disconnect one-by-one, but hell, this is strange. I always thought it was some spike of current causing an interference, but everything else works (including the motors, which are still running). Tonight I noticed one case where Autoslew was locked and gave up trying to connect to the mount, while the mount was happily tracking for some 15 minutes before I noticed it :P

 

Thanks

Link to comment
Share on other sites

  • 0

Yes, try it with only the main mount USB connected (not the mount auxiliary hub and not the covers, focuser, etc) and reboot your PC and cycle the power on the mount.  If that works then add the other USB cables back one by one until you find the problem one.

 

If you identify the one that is causing the problem then try the following:

 

1. Check for any problems in Control Panel / Device Manager.

2. Check you have the latest USB driver from Microsoft.

3. Check if your PC manufacturer has a later USB driver.

4. Check you have the latest driver for your USB-to-COM converter.

5. Check you are using the latest version of the virtual COM port driver.

6. Try changing the USB cable in case the cable is at fault.

7. Check the cable connectors and the PC/mount USB sockets for dirt or corrosion.

8. If you are using an external hub in the connection try with a direct connection in case the hub is the problem.

9. If you have equipment using the ASA USB hub on the mount then temporarily connect it direct instead.

10. If all else fails try using a different PC/laptop

 

I had very similar symptoms (Autoslew reporting random dropping of the USB connection).  As I said, my problem turned out to be a virtual USB over WiFi connection.  In your case it might possibly be the USB-to-COM converter.

 

After that I have run out of ideas!

 

Nigel

Link to comment
Share on other sites

  • 0

Alrght, I tried a simplistic solution and reduced the autostart test current for the DEC axis, and now it powers up right. When I increase the power to about 0.6A it drops the connection already there...

I wonder if somebody at ASA could tell me if this is an anomaly they see often or not. The balance could be imprved slightly, if that's the reason, but I still cannot understand why the motors themselves can keep running without problems even if the connection to the PC drops.

 

EDIT: Alright, the lower the speed, the longer it works. As soon as I put the speed to 3,2°/sec it lost connection on the DEC axis. I will proceed as per Nigel's checklist and see what I can do.

Edited by lukepower
Link to comment
Share on other sites

  • 0

I suspect the problem is USB-related (sockets, cable, driver, hub or conflicting virtual COM ports).  It would solve a lot of problems if Autoslew and the DDM mounts offered a physical COM port option!

 

However there may also be a balance issue complicating things - Autoslew is very sensitive to balance.

 

Have you checked balance at different mount attitudes to see if it changes much?  If you have any off-axis equipment (e.g. a finder scope or a filter wheel mounted at an angle) you may need to add additional counter-weights to get correct balance at all attitudes.  For example, I have a rotator which means that the balance changes at different rotation angles because of the off-axis filter wheel; I had to add custom-designed weights to the camera/filter assembly to keep it in balance as it rotates.

 

I normally do the initial balancing with both the counterweight arm and OTA horizontal, then test it again at 45 degree increments for each axis.  Then I do a manual tune (with the counterweight arm vertical and the OTA pointing at the pole).

 

Nigel

Link to comment
Share on other sites

  • 0

Hi Nigel,

 

well, I invested alot of time to match the balance on most positions. Indeed, there were some off-axis issues (mainly because my tube is a combination of a cassegrain and a newton). In RA, it is straight perfect in every position. In DEC, it is a bit off (but not by much). What I cannot understand is that, even with the loss of USB communication, the DEC motors keeps running, so to me it looks to be more a communicative error than a balancing one (otherwise the motor should shut down, I would guess).

 

Let's see, ASA support was very quick and offered me a remote session for tomorrow morning. Hopefully we can fix this :)

Link to comment
Share on other sites

  • 0

Alright, after the remote session with ASA it seems clear that I have an unbalanced scope :)

Basically it looks like the USB connection stays open, but the internal electronics somehow experience a voltage drop when the unbalance gets too much to be handled.

 

I will have to check this on the next clear night (plenty of snow here right now), these asymmetric balance issues are not that obvious to see at first. I read here that somebody connected magnets with lead weights, has anybody an idea on how they did that? (considering that lead is practically not magnetic)

 

Thanks

Link to comment
Share on other sites

  • 0

Hi Lukas, I am the one who used a lead weight:

 

I found a steel supplier who could make a steel strip of magnetic stainless steel (ordinary stainless steel is not magnetic).

He curved the steel to fit the radius of the tube and I attached it to the tube with adhesive tape.

I melted lead and cast a cylinder by pouring the lead into a mold.

I bent the lead disc to the shape of my scope tube.

I then drilled 8 small holes in the lead and glued small neodymium magnets into the holes.

The magnets are very strong so I put a layer of felt over the disc. This allows the disc to be slid up and down the steel strip easily, and the magnets hold it firmly in place.

I use the lead weight to give accurate balance in Declination, because a movement of the tube in the tube rings of a millimetre or two is too much.

I also offset the steel strip because the camera filter wheel is off centre.

 

George

 

magadjust.jpg

Edited by GeorgeCarey
Link to comment
Share on other sites

  • 0

Hi George,

 

thank you for this great info, I really appreciate it! I will try to get some lead shipped to me, magnets are already here (like to play with them :) ). 

Still have to find a way to get some stripes of magnetic stainless steel, let's see.

 

Thank you!

Link to comment
Share on other sites

  • 0

Alright, following to my problems with the balancing of the scope (which I still experience), it seems that the connection drops increase with the time the mount is powered up (that is, it got power but even with motors off). I am slowly going crazy right now :) 

Yesterday evening I had to slightly tune the motor PID parameters, which resulted in a more stable connection, but after a few hours it dropped repeatedly in some of the position I tested before - strange. Even the automatic Autoslew software unplug event, at that point, cannot clear the error. In the windows device manager, the COM port associated with the mount reports "Error 10: Could not startup device". The only solution is to close AUtoslew and power cycle the mount.

 

I am open for ideas :)

Link to comment
Share on other sites

  • 0

It appears your problem is still getting correct balance at all attitudes.

 

Until you get as near perfect as possible balance with the OTA in all attitudes it is not really worth exploring other issues as you won't know whether it is the balance or something else causing the problem.

 

Once the balance is sorted (using the Autoslew balance function) then, if the problem persists, you can explore other possible problems. Probably tuning is the next thing to look at after balance.

 

As always, change only one thing at a time so you don't complicate the diagnosis.

Link to comment
Share on other sites

  • 0

Hi Nigel,

 

yea, I guess I am also way too near the practical limits of the mount with my setup. I checked today (during the day) the balance, and found it as good as it can get in most positions. Let's see what happens tonight. It is funny in a way, as it seems the longer it runs the more prone it gets to these errors.

Link to comment
Share on other sites

  • 0

Hi Waldemar,

 

failing electronics would be bad news, to return the mount I would have to take some muscle strengthener to get the OTA off of it (100kg) :P

I guess that it's a combination of long OTA and heavy weight, these mounts are great but sensible to balancing issues. I am now going to remove an additional mounting plate I have on one side of the OTA, let's see if that helps.

 

Thanks 

Link to comment
Share on other sites

  • 0

Hi Nigel,

 

well, the OTA is about 84kg (that is, just the tube with the mirrors). Add to this:

  • 4" focuser (about 2,5kg)
  • ASA OK3Z focuser (about 2kg)
  • ASA 4" Cassegrain corrector (heavy, would guess 3-4kg)
  • some spacers (maybe 2kg)
  • FLI CFW 5-7 + FLI Proline 16803 (nearly 5kg)
  • one 2kg weight to handle the OK3z imbalance

so, all in all we are approaching 100kg. I am sure my balance is not perfect, I am still waiting to get some custom-made weights designed to exactly counteract the radial imbalance of the OK3Z, that should help alot. This Cassegrain-Newton tube is cool but difficult tobalance properly.

Link to comment
Share on other sites

  • 0

Hi folks.  First post - and considering a purchase. 

 

Reading between the lines this would appear to be a brownout of supply voltage on low level control electronics (the USB controller) as a result of transient high current in the torque motor.  A strategic L/C filter somewhere should abate this I'd have thought (and if the pcb is such that additional L is not possible, perhaps just boosting a decoupling capacitor is the quick solution?

Link to comment
Share on other sites

  • 0

Hi

 

Sorry, by brownout I meant a momentary dip in voltage at a critical place in the circuit that caused either a watchdog to fire or the ordinarily reliable performance of the controller to go haywire.  Voltage may only go down below the critical level for a microsecond to induce this depending on the circuit sensitivity.

 

Whether the system recovers on its own is important, and often dependent on so many things!  I have had circuits that lock up with such a voltage drop, requiring a full power down to come back to life.  I've also had them where recovery is dependent on pulling out the USB lead and putting it back in again (and of course others where recovery happens by itself). 

 

EMC susceptibility tests usually pick these issues up, and in most cases, very local decoupling around the USB controller circuitry at design time eliminates the issue before customer release.

 

Of course, I may be barking completely up the wrong tree, but it truly does sound like the USB drop-outs are coincidental with high transient current demands elsewhere, and that, whilst balancing the mount may stop the problem on all occasions, power or ground routing may be the root cause (and that the problem may come back to bite as mounts bed in and friction changes over time).

Link to comment
Share on other sites

  • 0

Hi Roger,

 

well, it actually sounds like the problem I am having. Please correct me if I'm wrong, but if I understand this correctly then this could be addressed by changing the circuit/PCB design? My hope was more like "add a condensator somewhere and you're done" :)

Link to comment
Share on other sites

  • 0

I have not studied anything - I am simply a prospective customer having admired the design and principles of the mount at Astrofest last month.  I've never even seen the inside of the mount, let alone the PCB layout!

 

If one is really fortunate this might be soluble by a rerouting of wiring so that the power to the high current stuff does not interfere with the area of the PCB with USB controller on it.  One would need to examine the topography and circuits to know if this would help (or would be possible).

 

It is also possible that the high rate of change of current in the motors is the problem - and that the interference is magnetic.  A ferrite bead or two on the high power wires might help there just to take the edge off the d2I/dt2.

 

One thing is for sure, it is always worth isolating as far as possible in such designs the control circuits from the power circuits.  Good layout techniques are as important (if not more so) than the circuit design itself.

 

One question - maybe naïve:  When you lose USB comms, does it ruin the current image, or is that ok, but you just lose comms for the next one?  Put another way, can you unplug the USB connection for a short while and connect it back up again mid-exposure without causing problems?  If that's the case, it may be possible to design a comparatively cheap external board that goes between the PC and the mount that "sniffs" the USB traffic, and simply flicks a relay on and off when comms goes wrong, forcing the USB to retrain from first principles.  Of course if you're using the auxiliary USB hub that would be out of the question as camera capture would be interfered with (although to prove the point a camera connection could be taken back to the computer without going through the mount).

Link to comment
Share on other sites

  • 0

Hi Roger,

 

well, when the connection is lost the camera and all other stuff on the USB hub of the mount still work. I anyway have to power cycle the mount (or its electronics, whatever) to get it recognized again from Windows... That's the main problem. Unplugging and replugging the USB cable doesn't help.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...