USB dropping out when roll off roof opens and closes

[Paul Haese]

Originally Posted by gary

Hi Paul,

With regards the inverter used to control the motor on the shutter, does it happen to be
SV008iC5-1F manufactured by LS or is it a SV iE5 or some other make/model number?

Best Regards

Gary

Hi Gary, it is a LS SViE5. Why?

[gary]

Originally Posted by Paul Haese

Hi Gary, it is a LS SViE5. Why?

Hi Paul,

Just have to duck out right now for a few hours so I will respond more fully
tomorrow.

Do you happen to know if the loom from the inverter to the motor uses shielded cable?

Best Regards

Gary

[Paul Haese]

Originally Posted by gary

Hi Paul,

Just have to duck out right now for a few hours so I will respond more fully
tomorrow.

Do you happen to know if the loom from the inverter to the motor uses shielded cable?

Best Regards

Gary

I would say no. It is 6mm cable 5 wire loom cable supplied by scope dome.

The cable is plastic coated only with outer casing and inner casings.

[tlgerdes (Trevor)]

I remember year ago working with Barry to solve a noise related problem at one of my customers, similar thing, turn device A on device B stops working, device A and B not related.

Worked out that we a had a noise frequency harmonic in the power at some 19kHz when device A turned on, device B was running a data rate of 19200bps. Turning on device A was corrupting data on device B.

When we looked at the power with an oscilloscope we saw lots of noise harmonics at different at frequency mulitples from this device.

[Paul Haese]

Some developments - today I tested a few things out. I ran the camera and Mount Hub Pro (MHP) on a battery and the problem still happened.

I replaced the existing cabling with a shielded power cable. Shielding not connected to anything. Not sure where to connect that. Camera still dropping out.

Then I ran the STL11000 and nothing happened. No drop out at all. Roof opening and shutting with no issues. This confirmed it had to be the camera and its USB connection.

I then replaced the MHP with my spare and this did the same thing with both cameras ie QSI dropping out and STL11000 not dropping out. Now I knew it must be the MHP not being able to supply enough power to its USB's or to power the camera.

So I remove the QSI from the MHP and provide the QSI with its own power and USB. It works without drop out. All the other gear is still connected to the MHP. Nothing drops out.

So it appears that I need to take all the scopes off the mount and put in another USB cable through the mount and provide power in the same manner to power the camera. Further testing will identify any further issues. So it seems that in the end it is the regulated power supply suffering from a slight voltage drop when the roof opens and this affects the power to the MHP and that in turn affects the camera and forces it to drop out.

Anything else I should check?

[gary]

Originally Posted by Paul Haese

I would say no. It is 6mm cable 5 wire loom cable supplied by scope dome.

The cable is plastic coated only with outer casing and inner casings.

Hi Paul,

The reason I asked about the model of inverter and whether the cable from the inverter
to the motor was shielded was because I had been thinking about your problem,
looked at the schematics from the ScopeDome web site, saw what models of
inverters they were using and then hunted down the User Manuals for them.

Here is a link to a web site in South Africa that happens to have a User Manual
for the SV-iE5.

http://www.google.com.au/url?sa=t&rc...49784469,d.dGI

You might like to confirm that this is the same model as yours.

If you could please turn to page 121 of 127 of the User Manual, I will attempt to
take you through it step by step.

On page 121 of the PDF, the manufacturer has their signed Declaration of Conformity
that they comply with EN 61800-3:2004 which is a standard addressing the
electromagnetic conformance (EMC) requirements of electrical power drive systems
and to EN 50178:1997, which is a standard addressing electronic equipment in
power installations.

Now it is important to appreciate that this Declaration of Conformity is for the
inverter only.

On the following page, 122, the manufacturer lists the applicable standards to
comply with the essential requirements of the European Electromagnetic Compatibility
Directive.

In other words, it's an engineer's way of saying "for this equipment to be used
in an installation where it is attached to a motor, these are the applicable
essential requirements".

If you could then please turn to page 123 of 127 in the PDF which has the title "EMI/RFI
POWER LINE FILTERS".

EMI stands for Electromagnetic Interference and RFI stands for Radio Frequency
Interference.

To quote the page -

THE LS RANGE OF POWER LINE FILTERS FF ( Footprint ) - FE ( Standard ) SERIES, HAVE BEEN SPECIFICALLY DESIGNED WITH HIGH
FREQUENCY LG INVERTERS.
THE USE OF LS FILTERS, WITH THE INSTALLATION ADVICE OVERLEAF HELP TO ENSURE TROUBLE FREE USE ALONG SIDE SENSITIVE
DEVICES AND COMPLIANCE TO CONDUCTED EMISSION AND IMMUNITY STANDARDS TO EN 50081

EN 50081 is the generic emissions standard. For equipment to be sold or
operated within the EU, they must by law meet this standard and a host of
others. Similar standards exist here in Australia.

What the text in this box is saying is "there are some power line filters that you
better use with this inverter to comply with the emissions standards and have less
trouble with other equipment nearby that might otherwise be affected by it".

They talk about the FF Series (Footprint) and FE Series (Standard) filters,
which are separate products they supply.

I refer you to the drawing at the bottom of page 123. They have separate
drawings for the FF Series and FE Series.

I further draw your attention to a few things on these drawings. Notice on both
drawings they have included a symbol for a common-mode choke on the cable
that runs between the inverter and the motor. This is precisely the device I was
recommending in my previous posts. Secondly, note the use of shielded cable
on that same cable. Thirdly, with the top drawing they have the FF Filter sitting
beneath the inverter (i.e. its footprint) but the dashed lines surrounding the
inverter, filter and choke I interpret to be an extra shielded enclosure.
Fourthly. notice that they show the inverter and filter to sit on an extended
metal plate which is to be connected to protective earth. Above the drawings they
have additional notes on the recommended installation instructions for the filters.

On page 124, they list the various model numbers of filters for the various models
of inverters and also have a mechanical drawing.

Finally, what chance you were supplied with one of these suitable filters and
common-mode chokes?

By the way, I noticed the inverter has a large number of parameters that can
be programmed and the front panel gives me the impression that one can
manually control the motor from it. Have you tried removing the cables
that interface the "USB box" to it and manually operated the motor from the
front panel to see if the problem still manifests itself? By pulling those cables,
you also reduce the loop area of one possible common-mode path.

In any case, my having pointed the bone earlier at the inverter-motor connection
dovetails with the manufacturer's own recommendations, but only time will
tell if my intuition was right. In any case, adding the filter, choke and shielded
cable sounds a lots easier than digging that trench.

I hope this advice proves helpful and solves the problem.

Best Regards

Gary Kopff
Mt Kuring-Gai NSW
02 9457 9049

[Paul Haese]

Thanks for the detailed response Gary.

I did not get any chokes or filters.

I have read the manual as you pointed out the sections and I find it odd that Chokes have not been supplied. I will keep the shielded cable. Do I install the shield at the earth point? The extended metal plates are heat sinks. I have earths on these as shown on the drawings.

I will place the shielded cable well away from the mains delivery cables.

Any suggestions as to chokes. The shielded cable needs to be wrapped twice through a ferrite core. What am I looking for here?

I take it the filters are like those you suggested earlier?

[h0ughy (David)]

Originally Posted by Paul Haese

Some developments - today I tested a few things out. I ran the camera and Mount Hub Pro (MHP) on a battery and the problem still happened.

I replaced the existing cabling with a shielded power cable. Shielding not connected to anything. Not sure where to connect that. Camera still dropping out.

Then I ran the STL11000 and nothing happened. No drop out at all. Roof opening and shutting with no issues. This confirmed it had to be the camera and its USB connection.

I then replaced the MHP with my spare and this did the same thing with both cameras ie QSI dropping out and STL11000 not dropping out. Now I knew it must be the MHP not being able to supply enough power to its USB's or to power the camera.

So I remove the QSI from the MHP and provide the QSI with its own power and USB. It works without drop out. All the other gear is still connected to the MHP. Nothing drops out.

So it appears that I need to take all the scopes off the mount and put in another USB cable through the mount and provide power in the same manner to power the camera. Further testing will identify any further issues. So it seems that in the end it is the regulated power supply suffering from a slight voltage drop when the roof opens and this affects the power to the MHP and that in turn affects the camera and forces it to drop out.

Anything else I should check?

glad you found it - i really consider the powered USB hubs as only an extension to get the USB past a few meters of cable extension. still recommend running two

[Paul Haese]

Originally Posted by Paul Haese

Any suggestions as to chokes. The shielded cable needs to be wrapped twice through a ferrite core. What am I looking for here?

I found these, just have to find a place locally that will sell these.

[cfranks (Charles)]

Paul,

Jaycar (in Adelaide) sell the Ferrite Rings and I think the largest is about 35mm ID. They used to sell a split one that you could clamp over an existing cable but it's not listed now.

Charles

[gary]

Originally Posted by Paul Haese

Thanks for the detailed response Gary.

I did not get any chokes or filters.

I have read the manual as you pointed out the sections and I find it odd that Chokes have not been supplied. I will keep the shielded cable. Do I install the shield at the earth point? The extended metal plates are heat sinks. I have earths on these as shown on the drawings.

I will place the shielded cable well away from the mains delivery cables.

Any suggestions as to chokes. The shielded cable needs to be wrapped twice through a ferrite core. What am I looking for here?

I take it the filters are like those you suggested earlier?

Hi Paul,

I suspected they hadn't supplied you with the line filter or choke.

On page 11 of 127 in the inverter manual, the lower drawing shows the cover removed
and they indicate a ground terminal. It is hard for me to tell from the line art if
that terminal is on the actual metal chassis of the inverter, but if it is, that is
where you want to connect the shield of the cable that goes from the inverter to
the motor. Use the shortest, fattest piece of cable you can find that allows
for attachment on the terminal and shield. Alternatively, if the cable is longer than you need,
peel back the shield, connect that to the ground terminal and cut short the wires
so that they remain shielded along their entire length.

If you had the inverter installed inside another metal cabinet or enclosure,
what they are suggesting is that the cable shield would terminate at a gland
on the enclosure at the point where the cables pass through into it.

Try and keep the entire cable as short as practical.

For that extended metal plate that you have identified as a heat sink, when you say
"I have earths on these as shown on the drawings", good, but could you please explain
to me exactly what you did there?

The ferrite toroid cores come in a range of dimensions. Here is a drawing of
one -
http://www.farnell.com/datasheets/1696282.pdf
This happens to be this one -
http://au.element14.com/wurth-elektr...8?Ntt=74270097
I happen to have purchased a couple of these myself in the last two weeks to use for measuring tiny currents.
However, if you search on the Element 14 web site, you should have no problem finding one
with an inner diameter that you think you will get the cable to loop through
a couple of times.

When you wrap the cable through the ferrite twice, if possible, try and keep
the two windings physically spaced apart. This helps reduce interwinding
capacitance, which is a good thing.

The manufacturer suggests placing the ferrite core as close to the inverter as possible.
The ferrite itself is a poor conductor but it has a high magnetic permeability.

What the ferrite does is it will choke common-mode currents passing through
the cable. The ferrite core can saturate so it has an upper limit to how effective
it is, but it will knock several dB of noise off.

Notice that the filter they recommend is on the power line side of the inverter.
They are trying to block noise being conducted back through the mains and into
your other gear.

Best Regards

Gary

[Paul Haese]

Thanks Charles, will take a look.

Gary, the base of the heat sink is the fixing point for the ground or earth cable. The instructions for fixing this are outlined in the Scope Dome manual. It says to fix at that point.

I will get a choke today and fix the tag of the shielded cable to ground point on the inverter. The filter is another question though. I will have to look around for one of those.

Thanks for the help.

[gary]

Originally Posted by Paul Haese

Gary, the base of the heat sink is the fixing point for the ground or earth cable. The instructions for fixing this are outlined in the Scope Dome manual. It says to fix at that point.

Hi Paul,

Yes, very good. Appreciate that. However, as grounding systems are usually a matter of compromises, could
you please paint me a picture with just a few words as to where specifically
the cable went to reference ground and roughly how long was the cable? For example, did you strap it to
the protective earth (PE) connection on the motor side of the inverter or all
the way to the ScopeDome USB box or to somewhere else?

I will get a choke today and fix the tag of the shielded cable to ground point on the inverter. The filter is another question though. I will have to look around for one of those.

When you Google for the specific model of line filter the manufacturer recommends
you don't get a lot of hits so it will probably coming down to finding another
brand at the right performance/price point.

Your post yesterday whilst I was composing mine gave rise to the possibility of
a power integrity issue.

On the regulated power supply you mentioned that was powering
lots of other things like your cameras, is it a fancier one with a built-in volt
and ammeter or have you happened to measure the typical DC current draw
with a multimeter that the appliances connected to it were drawing?

Normally you would want the regulated power supply to be delivering
a steady steam of power from its own built-in bulk capacitance to ride out
any dips in the mains. In turn, each digital appliance should ideally be powering
itself from its own decoupling capacitors sprinkled across its own circuit board
and those being recharged by the board's own bulk capacitors. But if you
have a lot of thirsty horses drinking from the same shallow watering hole all
at once, it could deplete very quickly and with it your noise margins will
also drop, leaving them more vulnerable to any noise from, in this case,
the inverter-motor pair.

As you will be aware by looking at the inverter user manual, it has a very large
number of parameters that can be programmed and monitored. Some of them can
control attributes such as the acceleration and deceleration ramps.
As at start-up the motor will look like a short circuit and draw the highest current,
there might also be the possibility of re-programming some of the parameters
in the inverter via its front panel to make the inverter/motor pair draw less when
it starts and stops. You can even change the acceleration and deceleration
ramps to S-shaped and so on. In other words, there might be some parameters
that you can change in addition to the other steps you are taking in order
to mitigate the risk that that the inverter/motor disrupts the operation of something
else or induces noise into your imaging.

Once again, hope this is helpful and good luck.

Best Regards

Gary

[icytailmark (Mark)]

sounds like its either the cables or the hub.try turning off everything and eliminate one device at a time paul. Hope this helps.

[Paul Haese]

Gary, the earth from the main board is screwed into the recess of the inverter where the earthing screw is located. That screw is in direct contact with the heat sink which is supposed to be the earthing point. That cable is about 3 meters long. Then the earth travels back to the power point and back up to the house which is about 25 meters away.

I don't know if the power supply is providing a perfect supply, but that might just be part of the problem. Below is a link to the power supply I bought. http://www.jaycar.com.au/productView.asp?ID=MP3096

The version I have is the 20amp supply.

Today I had a crash of the systems. The ST-I guide camera would not connect, not even the spare and the focusor would not connect either (but that is controlled by the mount hub pro). The main camera though is now connecting, but that has its own USB now. Next step is to go picking off each item until I have it sorted. I'll report back if I find what is going on here. Frustration is nothing compared to what I am feeling at present.

[Bassnut (Fred)]

I found the STi very unstable on a USB hub, usually wouldnt connect and if it did often droped out minuites later. With its own USB to the PC its fine.

[gary]

Originally Posted by Paul Haese

Gary, the earth from the main board is screwed into the recess of the inverter where the earthing screw is located. That screw is in direct contact with the heat sink which is supposed to be the earthing point. That cable is about 3 meters long. Then the earth travels back to the power point and back up to the house which is about 25 meters away.

Hi Paul,

I have in my mind the first iteration of how the inverter/motor system grounding might
be more ideally engineered and time permitting I will sketch it for you.
This then doesn't mean you should necessarily implement it in the way I suggest.
As I mentioned, grounding systems always require compromise somewhere.
However, hopefully it will illustrate some better practices that you can borrow from
if the other remedial steps you are taking still aren't sufficient for reliable
operation of equipment in your observatory.

Take heart that the term "ground" is one of the most misunderstood terms
in electrical engineering. Electrical Engineers working in different disciplines
within a system will use the term "ground" to mean different things.
So confusing it can be that I think if we could turn back time we would
happily purge the word altogether.

Cartoon attached below attributed to Bruce R. Archambeault, a Distinguished Engineer at IBM.

I don't know if the power supply is providing a perfect supply, but that might just be part of the problem. Below is a link to the power supply I bought. http://www.jaycar.com.au/productView.asp?ID=MP3096

The version I have is the 20amp supply.

Pretty hefty, but if you ever get the chance to use your multimeter to measure
the actual current draw of the gear you have connected to it, it might help
eliminate any surprises there.

Today I had a crash of the systems. The ST-I guide camera would not connect, not even the spare and the focusor would not connect either (but that is controlled by the mount hub pro). The main camera though is now connecting, but that has its own USB now. Next step is to go picking off each item until I have it sorted. I'll report back if I find what is going on here. Frustration is nothing compared to what I am feeling at present.

Sounds like Fred's anecdotal advice with his camera and the hub is good advice
and all you can keep doing is knocking these off one by one. There is always
a chance that more than one thing was going on here and in combination they
were enough to drop the overall system reliability.

Best regards

Gary

[Paul Haese]

Thanks guys, reinitialised everything today and all connected fine. Odd behaviour but imaging runs will tell me what if anything is going to drop out now. All things being equal I hope that this means the system is ok now. Yesterday I did install a heap of drivers again and but I did not fully test things at the end of the day as I wanted to get home. Today all seems fine.

Fred, the ST-I has worked fine for me all along through the hub. I have used it for months on end and never missed a beat, that was what was so annoying. All connected now, so I am expecting no other issues with it now. Perhaps I just did not reboot properly (just a simple restart instead of cold boot). Time will tell of course.

Computer is now running and collecting cloud data so I can revise the cloud sensor settings.

[Paul Haese]

Imaging run tonight. With the roof incorporated into CCDautopilot I get glichy stuff happening. Roof will open and close ok and camera is fine but focusmax might play up or pinpoint will not solve. Odd stuff. Take the dome software out of the equation and things seems smoother. Need to investigate further.

[Marke (Mark)]

Originally Posted by Paul Haese

Imaging run tonight. With the roof incorporated into CCDautopilot I get glichy stuff happening. Roof will open and close ok and camera is fine but focusmax might play up or pinpoint will not solve. Odd stuff. Take the dome software out of the equation and things seems smoother. Need to investigate further.

Just a thought but there maybe a port conflict with the dome driver and other software in that when called up its trying to access an assigned port
or address that is being used by something else .