All Sky Camera Revisited

[Gina]

No smoke - just not working - not detected.  I'll take it apart again and see if I can see what's wrong.

[DaveL59]

would you not be better to order a replacement socket:
https://uk.rs-online.com/web/p/type-b-usb-connectors/7480905/

[Gina]

I don't think I'd get the old one off without damaging the board.  The metal body is soldered on with two huge lugs and then there's 9 thin pins

[DaveL59]

Should be able to desolder with wick and a sucker and a little care, then you can get to the chassis pins to free them off and hopefully ease the old plug away. Fiddly with the number of pins but possible I'd think. 

Shame you can't just snip those 2 lugs from the other side and then tear the innards out to make extracting the pins easier

Edited May 5, 2020 by DaveL59

[Gina]

I've connected the camera with a USB3 cable to a USB3 port on my desktop machine and the USB is recognised as shown by lsusb in Terminal.  I thought when I had the problem with the ASI120MC-S that I ran an image capture software but now can't find what I did.  This is in Linux Mint.

[Gina]

I've connected the camera to the RPi using the USB3 cable and it's working so the USB2 contacts are working with this lead though they weren't working with a USB2 cable/plug.  This proves the camera is alright so I'll re-do my USB2 cable.

[Gina]

Change of plan regarding USB2 cable.  Drilled hole in side of camera case to take the cable with heatshrink sleeve to cover braid and avoid shorts.  A fairly tight fit.  Makes unshielded wires much shorter and no awkward bends in the cable.

[Gina]

Still doesn't work with that USB cable

[JamesF]

Continuity test on the cable?  Perhaps it's damaged somewhere...

James

[Gina]

Yes, thought of that and done it.

[JamesF]

Is there another ground pin that might need connecting up?  There are nine pins visible on the circuit board but only eight would be used for USB3.  I wonder if there's an expectation that the outside of the plug would be connected to signal ground as well and that's the ninth pin?  Bit of a wild guess, I have to admit.

James

[Gina]

That wouldn't agree with the USB2 spec though.  But I guess ZWO might not adhere strictly to the spec.

I'll try a separate Gnd.

Edited May 5, 2020 by Gina

[JamesF]

I can't recall what the spec says about the shield on the plug, I have to admit.  I'm a bit short on ideas otherwise though.  If the cables are all good and the core cable colours are consistent then the only other thing that occurs to me is that perhaps the cable is actually damaged, but still sound enough to conduct whilst not being good enough to carry a USB signal.

James

[Gina]

Connected a USB2 cable from the duff socket to the other RPi which has the shield connected to the common Gnd.  Still no joy.  I'm sure it's got to be the cable.  It's a new unused one so I don't know if it worked before.

[JamesF]

A new cable!  The horror!  I usually snip them off dead keyboards

James

[Gina]

I needed a right-angle A plug to go in the RPi and still fit in the box.

[Gina]

Seems I have another of those inexplicable problems to solve!! 

[symmetal]

Gina, I think your data & power connections are swapped over. On the USB3 connector datasheet the pcb footprint is shown from the component side. As you're viewing from the PCB side you need to swap the red and white wires over and as well as the black and green wires.

Alan

Edited May 5, 2020 by symmetal

[Gina]

Goodness gracious me - I believe you are right!!!  Thank you very much.  I'll swap them over.

[Gina]

That was it!!  It's now WORKING!! 

I can go to bed happy!

Edited May 5, 2020 by Gina

[symmetal]

23 minutes ago, Gina said:

That was it!!  It's now WORKING!! 

I can go to bed happy!

Yes, nothing worse than going to bed with a nagging problem on your mind.😀 At least your cable colours were correct. I wired up a USB cable I'd cut short and after it didn't work found that although the cable had the right colours they hadn't bothered wiring them to the correct pins on the moulded plug. 

Alan

 

[Gina]

I checked that the colours were correct with my DMM.  What I didn't check was that my brain was plugged in the right way round!! 😁

[Gina]

Though I'd lost the Peltier TEC I was going to use but have found it.  (I do have a smaller one in reserve but won't need that now.)   Just tested it by attaching it to a large heatsink with thermal grease and applying power.  Checked first with 5v then increased it.  With 10v applied for a while the cold side settled down to -8°C with the heatsink at 22°C.  It was quite happy with 13.8v applied, drawing 2.5A and reducing as the cold side got colder. 

As an experiment I tried a TEC1 - 12706 cheapo.  It took 4.5A and got hot both sides on 10v.  The thermal heating exceeded the Peltier cooling.  OK so the so called "cold" side wasn't as hot but...

[CedricTheBrave]

17 hours ago, Gina said:

Though I'd lost the Peltier TEC I was going to use but have found it.  (I do have a smaller one in reserve but won't need that now.)   Just tested it by attaching it to a large heatsink with thermal grease and applying power.  Checked first with 5v then increased it.  With 10v applied for a while the cold side settled down to -8°C with the heatsink at 22°C.  It was quite happy with 13.8v applied, drawing 2.5A and reducing as the cold side got colder. 

As an experiment I tried a TEC1 - 12706 cheapo.  It took 4.5A and got hot both sides on 10v.  The thermal heating exceeded the Peltier cooling.  OK so the so called "cold" side wasn't as hot but...

I found this with a cheapo one I had from Ebay. It melted the hot glue in the assembly and caused loads of damage. 

It's now in the bin

 

Where did you get the good one from and what model?

[Gina]

1 hour ago, CedricTheBrave said:

Where did you get the good one from and what model?

I'll have to check - the info should be in this thread (somewhere).