Loosing USB connection all the time

[lukepower]

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 ).

 

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

[Roger]

ok.  Well, I'd be interested to know from ASA if this is by design (it possibly is a safety feature - too much power being requested so lock out - in which case it may be a parameter they would be prepared to adjust), or an unintentional lock-up, perhaps resulting from the brownout, which I contend should be soluble.

 

I wonder if for expedience, turning down all the mount PID gains by, say, 20%, would help?  I must stress this is with no mount experience just clutching at straws.

 

Oh - last thought:  Is the power supply man enough or is it possible that is browning out?  How long are the leads to it?  Might be worth putting 10,000µF across the main rail connections up at the mount end of the power cable.

 

<small edit for clarification of capacitor position>

Edited March 2, 2014 by Roger

[lukepower]

Well, I use a Siemens PSU with 24V and 20A output, short leads (it's sitting at the base of the mount). FUnny enough, the motors keep often running even without connection to the PC

[Roger]

Well, I use a Siemens PSU with 24V and 20A output, short leads (it's sitting at the base of the mount). FUnny enough, the motors keep often running even without connection to the PC

 

In the FAQ it says that peak draw is 20A on slewing - so this could be the issue.  Please humour me and put a couple of car batteries in series there for a while (either alone or in parallel with your Siemens block) and see if the problem goes away.

 

EDIT - Best to use the batteries standalone for this test - not sure if that Siemens lump likes being put in parallel with batteries.

Edited March 2, 2014 by Roger

[Roger]

Or if you haver access to an oscilloscope see if you can detect any appreciable drop-out on slewing - may be easier.

[lukepower]

Hmmm, the PSU should also have  apeak of higher than 24V... Well, have to investigate this, too...

[Roger]

Folks - I have likely made about 5 by adding 2 and 2.  Apart from the safe experiments I have suggested of additional decoupling close to the mount power pins and replacing permanent power supplies for a short time with lead acid batteries, please do not go off on tangents without checking what you're doing with ASA.  I may be totally wrong with my thoughts.  I am just an engineer considering a purchase and have never seen one in action apart from for a brief 5 minutes last month.

Edited March 2, 2014 by Roger

[lukepower]

Hi Roger,

 

I do have a remote support session planned for tomorrow, and I will ask their (ASA) engineer what are his toughts on this. Too bad that there is no way of easily exposing the electronics of the mount (without disassembling everything). I could try with something like a 40A PSU: http://www.tpautomation.de/shop/Stromversorgungen/SITOP-Stromversorgungen/SITOP-modular/6EP1337-3BA00-SITOP-modular-Stromversorgung-1-2-ph-DC24V-40A::3557.html

 

The thing that disturbs me is that in fact we talk about loosing the communication channel to the mount, the mount itself kept working. Once I had it tracking for over half an hour before I saw that Autoslew lost the connection. A voltage drop might well be the reason why the electronics loose USB connectivity for a fraction of a second...

[Roger]

I like the look of that 40A power supply!

 

However, if as you say the mount keeps going, then the glitch internally is I guess not likely to be solved by a beefier supply - unless of course it actually stops and immediately recovers - which is not the same thing.  If you have one (or a couple of lead acid batteries) to hand, it is worth a try for sure.

[lukepower]

Let's see what the remote support tomorrow yields. I changed USB cables (several times), added an industrial grade USB hub (with filters), and countless trials of changing location of the cables running from the mount to the power supply and to the computer. Cross fingers and let's see...

[GeorgeCarey]

Have you tried using a different computer?

Perhaps it would be a good test to try.

[lukepower]

Hi George,

 

yes, I did. No change at all :/

By the way, I had a remote support session with ASA today, and they decided I should return the mount for a closer inspection. It looks like the electronics have something wrong with them...

[Waldemar]

Well, Lukas, at least a little bonus for all the body building workout you did in case of this event...

and... your mount will be checked, checked and double checked... they will leave nothing uninvestigated.

 

When I sent my mount in for upgrading with absolute encoders, I was really surprised when I got it back:

They did everything! Not just encoders, but wiring, new connectors, powered USB hub, electronic boards, the works!

 

Right now it may be a pain in the uhhh...  neck (literally), but you will be glad when it's done   and have your self confidence back.

 

 Waldemar

[lukepower]

Hi Waldemar,

 

heh, your post got me smiling I literally smelled it since a while, that I would have to return it. These moments I wish I had a crane or lifting system in the observatory to get the OTA down from the mount. But heck, at least I have an excuse to convince my +1 to go to the Steak house once again

 

I actually hope that the guys at ASA will give it a nice Upgrade (*cough* maybe they are reading this? *cough*) with stuff like 50A-motors and ethernet connection

I might add a few bottles of beer or wine to help out a bit

 

So for now, I'll just keep this beast of an OTA asleep, hopefully they find the flaw and can fix it

[Waldemar]

I am sure they will do a great job, Lukas.

You will have a lot of good times with your mount!

 

If I were you I would give them the wine afterwards... I mean...those Austrians, you know.

Hell, they make 80% proof rum, that must mean something... hahaha

 Alternative: come over and I will help you finish them :lol:

 

Waldemar

[lukepower]

Hehe, well, we do have enough bottles in the cellar to survive quite a bit of harsh times, so... Let's see. Once it works I am quite sure to have a party over here

[Waldemar]

You mean.... I am invited??? :D

[lukepower]

Oh well, when it works...

We will probabily feel the effect of "Grenzgrößenwasser" once we start with it, but who cares, the system is automatic and we can look at the pictures the day after

[Roger]

How many is this communication drop-out affecting?  I believe I have identified at least two on here.

[nakbrooks]

Bit difficult to say as USB problems can have so many causes. I had a similar drop-out issue but it turned out to be a USB over WiFi converter causing the problem - nothing to do with ASA.

 

I really think there is an argument in favour of providing a serial interface as an option as this is much more reliable because it is low tech.

Edited March 3, 2014 by nakbrooks

[Roger]

I would agree about RS232.  The problem is most PCs these days don't have a native serial port - and the USB converters decouple you from the real RTS/CTS and other control lines!

 

Let's hope it is a 1-off and there is no generic problem.

Edited March 4, 2014 by Roger

[nakbrooks]

Yes, but for high-end mounts (e.g. DDM85XL) designed primarily for fixed installations I think RS232 should be an available option as many people would use dedicated PCs that either have add-in serial cards or built-in serial ports (a dedicated PC with an add-in serial card can be got for only a few hundred pounds after all).  I use a wide temperature range industrial PC with 6 serial ports (none of which I am able to use as all the ASA kit uses USB).

 

I am a firm believer in the KISS principle (Keep It Simple, Stupid); there are enough variables involved in amateur astronomy as it is, without having to struggle with the hardware/software issues involved in getting USB to work reliably through daisy-chained hubs with different combinations of Windows/vendor drivers and multiple competing devices sharing the same USB bus.

[lukepower]

Hi Nigel,

 

agreed. USB is easy and all, but the troubles you can have with those stupid drivers and cables... RS232 or similar serial connections should be fast enough and stable. I personaly would also consider an Ethernet connection. My guess is anyway that the issues arise, at least in my case, between the mount electronics and the Serial-to-USB converter sitting in the mount, as only the connection is lost but the motors are still tracking...

[lukepower]

Addendum: I just came back from unmounting the OTA and mount. Once the OTA was away, the DEC axis had a very strong smell like burnt plastic. I removed the cover of the DEC axis (with those 4 small screws) and found a circuit under there with part of it completely burnt. Looks to me like a voltage regulator, but I might be wrong. Will post a picture afterwards.

So, at least I know it is a defective electronics which probabily is cause of the problem.

[GeorgeCarey]

Good to hear that you have found the problem!

[Roger]

I also agree regarding dedicated serial ports.  The present setup here is a PC with a 4-port serial card (PCI) installed (one for the focuser, one for the mount, one for the autoguider and one spare!).

 

Glad you've identified a burned out part.  However, is that cause or effect?  I hate to be the prophet of doom, but if you have an issue where the micro or other control circuitry is misbehaving for any length of time, this may be a symptom of the failure mode induced by a brownout rather than the cause of lost comms (its drive may have gone half level or similar allowing high current and high voltage at the same time across a switching element that ordinarily experiences one or the other).

 

I stress yet again I've not seen any circuits but I would find it incongruous that the logic / low level electronic supply lines would be sub-regulated from a high current motor drive rail (they would be regulated from a separate feed from the prime mover independently decoupled before any voltage reduction).