20 replies to this topic

[Lymelight]

I had an unfortunate experience with my Starlock. I had moved it to the side saddle and didn't inspect cable reach (I guess) because as it was slewing to M51 -- the cable must have hung up because it got sucked out of the connector on the Starlock side.

 

It was on High Precision and locking in on a star when this happened. Then the Handbox display blinked and the LX850 rebooted. The mount itself was fine after I did a quick Align again, reset my PARK position etc. It works fine each night since but I have to turn on HIGH PRECISION every time I start up, center the star by hand etc.

 

I repinned a connector, easy enough, but no power to Starlock.

 

I received a fresh cable from Nelson, plugged it in, powered up the LX850, Starlock--- nothing. No Starlock in the MENUs, like it is not there.

 

Mount works. Handbox works. But no Starlock. What to do?

[astropgr]

You may have loosened a connection internally on the starlock electronics board. OR you may have jumped a power pin during the ordeal and shorted the electronics in it. Can you open it up and check for connection issues or fried components? 

[Lymelight]

You may have loosened a connection internally on the starlock electronics board. OR you may have jumped a power pin during the ordeal and shorted the electronics in it. Can you open it up and check for connection issues or fried components? 

I suppose I can dismount it and open it, yes. What to look for and how, hmmm…

[Lymelight]

In anticipation of any add'l comments and guidance, I plan on testing continuity and power. I found an OzAndrew post on pinning:

 

"OK, i just probed my test bench with a pigtail in the port

With the DMM ground probe on "B" (yourBlue), i get

"A"(Yellow) = +12V

"C"(Green) = -8V

"D"(Red) = 0V

Andrew Johansen Melbourne Australia"

 

I am confident about the wiring itself, ie which pin on each side maps to which pin on the other side. Based on this, my plan is:

1. I will test my new [Nelson Devices] Starlock cable for A-->A, B--B, etc. Don't anticipate problems there but doing it.
2. I will snip and strip my damaged cable on the Starlock end, do the same. If I have continuity on all four leads...
3. I will plug that as a pigtail cable into the mount and do the OzAndrew checks. That would confirm power out into the cable.
4. If that checks out ok, I will plug the new cable into Starlock, pull the head off, and if I can, do the same tests inside Starlock on the other side of the pins. That would hopefully uncover a loose connection where the RJ pins sit.
5. If that checks out, then the damage is downstream of the power entering into the Starlock unit. That will exceed my pay grade.

Edited by Lymelight, 05 April 2024 - 10:13 AM.

[Lymelight]

Performed continuity checks on the Black to Black (12v) wire. Check.

 

Performed checks on each contact inside the Starlock head with the LX850 powered up, and the cable connected. If I was reading the solders correctly right-->left as corresponding to the pins 1-2-3-4, ie Yellow; Green; Red; Black:

 

Right/Yellow/Ground--- used as Ground for each probe below

Next Green/Tx From Starlock-- +8v 

Next Red/Rx from Mount--  negative DC voltage (?? this conflicts with OzAndrew's bench test)

Next Black/12V DC-- +12v

 

 

Not sure if the culprit here is the Rx From Mount, eg Pin 3, eg Red as cabled by Meade (eg "C" per OzAndrew test bench results, which he reported as "zero").

 

But power is running from the mount into the Starlock at Pin 4/Black wire. No Starlock light though, and no STARLOCK in the UTILITIES menu. So, next I am going to attempt to run the Starlock Utility on the RS232 port, since there is power somewhere in there. Maybe something in doing that will kick the Starlock awake.

Edited by Lymelight, 05 April 2024 - 01:27 PM.

[OzAndrewJ]

Gday Lymelight

next I am going to attempt to run the Starlock Utility on the RS232 port, since there is power somewhere in there.

The LX850 main board is identical to the LX200s except it uses a different cover plate stencil.

ie the Starlock port on the mount itself is just the second rs232 port underneath.

If you unplug StarLock and plug your std 507 cable into the StarLock port,

it will work like normal.

soooooo, yes, there is a +12V supply in there

Re my bench test, i do so many i cant remember what i did or when. ( So i re did it )

As i dont have a Starlock, i am assuming i was merely probing a pigtail in the mount itself????

If so, the way rs232 works is a -8V means logic 1 ( default ) and a +8V means logic 0

When when not connected to anything the mount holds its Tx line at -8V and the Rx line "floats".

When something "talks" to the mount, the Mount RX line will swing between +8v and -8v

The same would probably happen for the StarLock head that is remotely powered by a pigtail

ie its Tx would sit at -8V and its Rx line would float

When connected together, both the Rx and Tx should be asserted when its not talking

so neither should be at 0V, sooo based on

Not sure if the culprit here is the Rx From Mount,

Its not technically Rx from Mount, its Tx to Mount  ( as the mount just passively receives data )

and that would indicate the StarLock unit is not holding its Tx line at -8v

 

 

Andrew Johansen Melbourne Australia

Edited by OzAndrewJ, 05 April 2024 - 04:46 PM.

[Lymelight]

Gday Lymelight

The LX850 main board is identical to the LX200s except it uses a different cover plate stencil.

ie the Starlock port on the mount itself is just the second rs232 port underneath.

If you unplug StarLock and plug your std 507 cable into the StarLock port,

it will work like normal.

soooooo, yes, there is a +12V supply in there

Re my bench test, i do so many i cant remember what i did or when. ( So i re did it )

As i dont have a Starlock, i am assuming i was merely probing a pigtail in the mount itself????

If so, the way rs232 works is a -8V means logic 1 ( default ) and a +8V means logic 0

When when not connected to anything the mount holds its Tx line at -8V and the Rx line "floats".

When something "talks" to the mount, the Mount RX line will swing between +8v and -8v

The same would probably happen for the StarLock head that is remotely powered by a pigtail

ie its Tx would sit at -8V and its Rx line would float

When connected together, both the Rx and Tx should be asserted when its not talking

so neither should be at 0V

Andrew Johansen Melbourne Australia

Hello Andrew!

That is amazing... I followed that. Honestly I am simply blind when it comes to electronics. But I followed, and thank you. Yes, your old test was a pigtail in the mount. And here I was in the powered-on mount, but into the Starlock board, too, and so it makes sense that one was at +8v and the other (couldn't get there on this testers scale) was hunting for -8v, no doubt. That made a logic 1 and a logic 0. I suppose that's a little bit of good news--- taht's all in order. But that Starlock is out of commission.

 

I hooked it up to a laptop running the Starlock Utility. I believe it wanted COM3 to connect....that seemed to connect. The otehr 3 ports had a definitive Windows dialog box pop up, Starlock Not Found or something to that effect. COM3 never displayed that message. Beyond that, I never had anything but a black screen in daylight, or the RA Dec screen but no activity, or PSF would be 'hunting for a star'... almost like it couldn't see. And all that time, no red led on the back of the Starlock.

 

I fear the worst that during the tug-until-dislocated pull on the hung cable, there must have been a moment when something broke or got shorted out. But at the pins where the mount's leads come in, all works as it should.

[Lymelight]

Gday Lymelight

The LX850 main board is identical to the LX200s except it uses a different cover plate stencil.

ie the Starlock port on the mount itself is just the second rs232 port underneath.

If you unplug StarLock and plug your std 507 cable into the StarLock port,

it will work like normal.

soooooo, yes, there is a +12V supply in there

Re my bench test, i do so many i cant remember what i did or when. ( So i re did it )

As i dont have a Starlock, i am assuming i was merely probing a pigtail in the mount itself????

If so, the way rs232 works is a -8V means logic 1 ( default ) and a +8V means logic 0

When when not connected to anything the mount holds its Tx line at -8V and the Rx line "floats".

When something "talks" to the mount, the Mount RX line will swing between +8v and -8v

The same would probably happen for the StarLock head that is remotely powered by a pigtail

ie its Tx would sit at -8V and its Rx line would float

When connected together, both the Rx and Tx should be asserted when its not talking

so neither should be at 0V, sooo based on

Not sure if the culprit here is the Rx From Mount,

Its not technically Rx from Mount, its Tx to Mount  ( as the mount just passively receives data )

and that would indicate the StarLock unit is not holding its Tx line at -8v

 

 

Andrew Johansen Melbourne Australia

Add'l info, I just spoke with Sal at Meade. Bottom line, I ordered a replacement Starlock. Low risk... they don't have any to ship separately just yet. But I'm in the queue.

 

Here's why: to look at my Starlock, they want the Starlock, the hand controller, and the control unit off the LX 850. I question the necessity of that, of going out of commission for as long as it takes them to look at it. (In fact, possibly the same lead time as getting a replacement shipped.)

 

But I have a question about what swings me: if I had it cabled to a powered-up mount and at the Starlock card itself, I probed those pins and all four behaved correctly, does that mean by physics alone the issue has to be somewhere on the Starlock board itself? Or could it be in the control unit on the mount despite the +12v/+8v/-8v reported in real time?

 

And if it has to be on the Starlock board, I conclude I am best served getting in queue now for a replacement while I find out if this could even be serviceable anyway. If it can, reliably for half or less of a replacement, that could be the way to go. I'd just cancel the Meade order.

 

What do you make of that line of reasoning. Please, if I'm looking at it wrong pls say so.

 

From up here in cloudy, chilly Lyme CT USA,

but heading to the path of totality with a C5 and EQ-G on Sunday,

 

Thank you. Alan.

Edited by Lymelight, 05 April 2024 - 05:26 PM.

[OzAndrewJ]

Gday Alan

I have no detailed knowledge of the Starlock unit itself,

and AFAIK, no one has ever reverse engineered the schematics.

The StarLock unit is a fully functional computer unit that can have intelligent

conversations with the mount using serial commands.

ie when you connect your PC to the mount as per normal use,

only the PC ever "sends" commands and the mount responds.

With Starlock ( as it can guide and do all sorts of other things ),

there is a two way conversation, ie the Mount tells starlock what to do and when to do it

and as part of doing that, Starlock may then send commands back to the mount.

That is why Meade would also want the main board.

That said, if you have the ability to make a cable with tapping points for Gnd, Tx and Rx

and access to a COM port sniffer, ( or CRO ) you could simply probe the Tx and Rx lines during booting

to see if the handshaking between the Mount and StarLock is happening.

That would immediately indicate if the actual comms is working.

 

Andrew Johansen Melbourne Australia

[Lymelight]

Gday Alan

I have no detailed knowledge of the Starlock unit itself,

and AFAIK, no one has ever reverse engineered the schematics.

The StarLock unit is a fully functional computer unit that can have intelligent

conversations with the mount using serial commands.

ie when you connect your PC to the mount as per normal use,

only the PC ever "sends" commands and the mount responds.

With Starlock ( as it can guide and do all sorts of other things ),

there is a two way conversation, ie the Mount tells starlock what to do and when to do it

and as part of doing that, Starlock may then send commands back to the mount.

That is why Meade would also want the main board.

That said, if you have the ability to make a cable with tapping points for Gnd, Tx and Rx

and access to a COM port sniffer, ( or CRO ) you could simply probe the Tx and Rx lines during booting

to see if the handshaking between the Mount and StarLock is happening.

That would immediately indicate if the actual comms is working.

 

Andrew Johansen Melbourne Australia

It's getting deep now. I don't know about COM port sniffers. If I may, I wonder if the Starlock Utility was telling me anything definitive that the mount is not the issue, I don't know. It is said that to use the Utility, the Starlock must be powered up by the mount.

 

My heart is sinking now thinking I could indeed be sans anything more than locking and unlocking clutch screws. That would make for a very cumbersome 'slew and hunt in the finderscope' setup. I'd be relegated to the 5" on an EQ-G for as long as Meade takes.

 

But I might have to bite the Plan B bullet and send it all in after all. I'll check on Plan B Monday with them. A very impactful cable snag.

[OzAndrewJ]

Gday Alan

I cant help re Starlock or its utility, as i have never seen or used a real Starlock.

That said, if you can make cables, what you can probably do is make a short interconnect

that goes between the Starlock port and the rs232 port.

It only needs 2 wires and connects the Tx of one plug to the Rx of the other and vice versa.

As the board in the LX850 is based on the LX200s, it has the same safeload code in it ( so far )

If so, you can start the mount in 444 mode ( like safeload but it is for mount diagnostics )

It has a series of tests for all the motors, sensors, focusser, Leds, GPS, comms etc,

so will probably not work for the sensors ( as they are different to the LX200s )

but it can test the motors, comms etc.

In the com test,

the mount sends 256 bytes out of port 1 and receives it on port 2

then reverses, ie sends from port 2 to port 1

If that passes, you know the mounts comms and panel connectors are OK

so could tell Meade that, and then only have to send in the Starlock unit??

( as they can then test with any panel )

 

Andrew Johansen Melbourne Australia

[Lymelight]

Gday Alan

I cant help re Starlock or its utility, as i have never seen or used a real Starlock.

That said, if you can make cables, what you can probably do is make a short interconnect

that goes between the Starlock port and the rs232 port.

It only needs 2 wires and connects the Tx of one plug to the Rx of the other and vice versa.

As the board in the LX850 is based on the LX200s, it has the same safeload code in it ( so far )

If so, you can start the mount in 444 mode ( like safeload but it is for mount diagnostics )

It has a series of tests for all the motors, sensors, focusser, Leds, GPS, comms etc,

so will probably not work for the sensors ( as they are different to the LX200s )

but it can test the motors, comms etc.

In the com test,

the mount sends 256 bytes out of port 1 and receives it on port 2

then reverses, ie sends from port 2 to port 1

If that passes, you know the mounts comms and panel connectors are OK

so could tell Meade that, and then only have to send in the Starlock unit??

( as they can then test with any panel )

 

Andrew Johansen Melbourne Australia

I just ran the 444 test, and it reported SERIAL 00 FAIL. But from another post, does it need to be cabled in a particular way? Can I run it with the Meade serial cable, eg, while Autostar Suite is running? 

[Lymelight]

I just ran the 444 test, and it reported SERIAL 00 FAIL. But from another post, does it need to be cabled in a particular way? Can I run it with the Meade serial cable, eg, while Autostar Suite is running? 

 

Gday Alan

I cant help re Starlock or its utility, as i have never seen or used a real Starlock.

That said, if you can make cables, what you can probably do is make a short interconnect

that goes between the Starlock port and the rs232 port.

It only needs 2 wires and connects the Tx of one plug to the Rx of the other and vice versa.

As the board in the LX850 is based on the LX200s, it has the same safeload code in it ( so far )

If so, you can start the mount in 444 mode ( like safeload but it is for mount diagnostics )

It has a series of tests for all the motors, sensors, focusser, Leds, GPS, comms etc,

so will probably not work for the sensors ( as they are different to the LX200s )

but it can test the motors, comms etc.

In the com test,

the mount sends 256 bytes out of port 1 and receives it on port 2

then reverses, ie sends from port 2 to port 1

If that passes, you know the mounts comms and panel connectors are OK

so could tell Meade that, and then only have to send in the Starlock unit??

( as they can then test with any panel )

 

Andrew Johansen Melbourne Australia

The loopback cable:

1- That would be a short cable between RS232 and Starlock ports, both on the LX850 panel?

2- Would I pin 2-->2 and 3-->3, or cross them over, 2-->3 and 3-->2?

 

If I understand well, correctly wired, during the 444 test it would show that the mount itself is pinging and receiving within its board.

[OzAndrewJ]

Gday Alan

Yep, you need a loopback cable between the 2 ports on the control panel

It only needs 2 wires so that   

Port1 Tx  goes to Port2 Rx and

Port2 Tx  goes to Port1 Rx

 

Andrew Johansen Melbourne Australia

 

[Lymelight]

Gday Alan

Yep, you need a loopback cable between the 2 ports on the control panel

It only needs 2 wires so that   

Port1 Tx  goes to Port2 Rx and

Port2 Tx  goes to Port1 Rx

 

Andrew Johansen Melbourne Australia

 

LX200 LoopBack Tester.jpg

Made one, ran the 444 and voila Serial PASSED. Ran up to FF and Passed.

[OzAndrewJ]

Gday Alan

OK, based on what you have said so far, we know the main panel is supplying +12V to the Starlock unit

and the Main panel comms are working as expected.

As such, it is 99.937% sure to be a fault in the Starlock unit itself.

Maybe talk to Meade and tell them you have done all these tests

and they wont make you send the panel in.

At least you can keep using the mount in the interim.

Andrew Johansen Melbourne Australia

[Lymelight]

Gday Alan

OK, based on what you have said so far, we know the main panel is supplying +12V to the Starlock unit

and the Main panel comms are working as expected.

As such, it is 99.937% sure to be a fault in the Starlock unit itself.

Maybe talk to Meade and tell them you have done all these tests

and they wont make you send the panel in.

At least you can keep using the mount in the interim.

Andrew Johansen Melbourne Australia

Appreciate that Andrew. I have sent them the updates about Autostar Suite moving the mount from either port, and that the Serial PASSed, as well as that we couldn’t get the StarLock to talk to a working LX200. They did say today they could take the Starlock-only ‘at my risk’, but at the quoted estimate and given their turnaround of 6-8 weeks and given all the indicators that the Starlock is the issue—not sure I’d even bother sending it in because if they find it defective on a repair attempt, I don’t think they repair the boards, they replace them, and I have one on order already. So, I’ll keep using the mount in the interim. I don’t have any big issues beyond having to center stars with the handbox for High Precision slewing (and I think their PEC and ARC routines all lean on the Starlock to do their work, too, so none of that for now, either).

[OzAndrewJ]

Gday Alan

I am 100% sure they wil not do any "repairs" as its not worth it

They would start it up on the bench and if it doesnt work, just dump it.

That said, doing the ARC relies on Starlock but its non critical, 66% wil work OK

If you have already loaded a PEC model, it will also continue to work

and if not, you can stil use the std methods to create one,

just you need to guide vs Starlock.

ie The stored model is the same, just the creation method changes

Andrew Johansen Melbourne Australia

[Lymelight]

Gday Alan

I am 100% sure they wil not do any "repairs" as its not worth it

They would start it up on the bench and if it doesnt work, just dump it.

That said, doing the ARC relies on Starlock but its non critical, 66% wil work OK

If you have already loaded a PEC model, it will also continue to work

and if not, you can stil use the std methods to create one,

just you need to guide vs Starlock.

ie The stored model is the same, just the creation method changes

Andrew Johansen Melbourne Australia

Yes exactly on the repairs. Best to spend my money on the replacement and move on. 
 

I do believe flashing firmware would have wiped out my PEC. My User Objects were gone, and I’m just assuming the PEC is as well. But I’m not concerned. I’ve focused on the basics while the Starlock is down.

 

1- I obtained a 14” Tribahtinov mask and it has been so helpful. Now I'm getting avg star sizes consistently under 3". Was not the case. I love it so much I got one for the C5, too. It's a wonder tool.

2- No Starlock means unaided tracking. Ran a PA routine in ASIAIR Plus last night. It reported 5' error off NCP. Starlock compensated for the error hence Autostar Autodrift reported no adj necessary-- because it knows it can compensate. But now with no Starlock, in fact it needed a tweak. So I ran the routine and moved az and alt as instructed. I am now 4" off NCP. 4 arc-seconds. 0.1" in elevation! The rest comes just from tightening the bolts.

 

Example, here is last night’s effort with a bit of wind and a rising, nearly full moon. 40’ of subs, reduced to 30’ kicking out the windy frames. 
 

 

Constantly experimenting and learning.

[OzAndrewJ]

Gday Alan

I do believe flashing firmware would have wiped out my PEC. My User Objects were gone, and I’m just assuming the PEC is as well.

I am pretty sure your user data doesnt get overwritten by a load.

Its easy to check if you have patched data as its all stil in there.

If you use my PEC editor, it can show the currently stored data

and also recover the historical data ( tho its a bit slower )

Andrew Johansen Melbourne Australia

[Lymelight]

Gday Alan

I am pretty sure your user data doesnt get overwritten by a load.

Its easy to check if you have patched data as its all stil in there.

If you use my PEC editor, it can show the currently stored data

and also recover the historical data ( tho its a bit slower )

Andrew Johansen Melbourne Australia

It's academic, as the comets in question have moved on, but my user objects are definitely gone so I only assume that's from the firmware update that was flashed. I can't say for sure as I had someone else doing that as part of troubleshooting; maybe something else also affected the user objects. As I say, only academic but your conviction is noted. What's more important: Now within 4 arc-seconds of the NCP sitting on a steel pier mounted on 3500 lbs of cement below the frost line. Not that it was an issue before, but geez it will be interesting to see if that 4" shifts over time. 

 

Edited by Lymelight, 25 April 2024 - 05:45 PM.