Power Supply

[kermit]

HI,
I have had some power supply issues (mount behaving funny when autoguiding) and I was wondering what power supply you use and how you connect everything together. I have been told that the issue with the mount could be linked to a power supply issue.
The power supply to the observatory has been installed by a professional electrician. I now have a sub panel with 8 sockets all attached to the same earth. The earth is a new earth fitted close to the observatory and has nothing to do with the house one.
I need to power the following:
1 Mount fitted with the Gemini 1
1 Imaging Camera
1 Guiding camera
1 Usb powered hub
3 Dew bands
1 Robofocus
What would be the best way to power the lot?
Thank you for your help.
Vince

[Waxing_Gibbous]

I had similar problems arising from using the 2-phase power from the house. Every time the freezer kicked in or the kettle tuned on, the voltage to the mount would drop.
Now I use a Celestron power tank in "stand alone" mode and that's fixed it.
You might need two or three to run all that gear, but the delivery would be uninterrupted.
However, check with your sparky as I know less about electricity than almost any other subject.

[The_bluester]

For my eventual pier and obs setup I have been thinking about a bench DC power supply, potentially fed via a UPS on the AC side so it does not drop if there is a mains interruption.

The UPS could potentially supply any AC fed gear at the same time to keep everything up and running but for things like a laptop that would be a bit belt and braces as it would just drop to battery anyway.

The other way I could see to do it would be the bench supply but with moderate sized SLA battery across the output to pad through any interruptions, the problem with that would be the need for the supply to be on constantly to keep the battery trickle charged, but then that is what an UPS would be doing as well.

In my case it is probably overkill, given that I am purely a visual observer a brief loss of power is an inconvenience more than anything else, at worst it would result in having to redo a star alignment on a goto mount.

[kermit]

Thank you for the advise it sounds pretty good.
Vince

[wasyoungonce]

Vince. I have seen & been following your discussions on yahoo Gemini forum re: autoguiding issue. I didn't want to interfere with the discussion with Tom Hilton. Tom knows more of the Gemini and has forgotten more than I will ever know, but I am sometimes useful on Gemini 1 repair matters.

At this stage all I can say is to remove all non essential accessories (just have Gemini and autoguider powered up, indeed power the autoguider from the Gemini spare power port if necessary) and try a straight +12V battery supply, no transformer or switchmode PSUs!

Your problem is perplexing and indeed it has me confused. However there is a method to test the ST4 autoguide port. I (and Tom Hilton, the most helpful person of Gemini fame) have built ST4 guide port testers. They consist of just 4 momentary switches that switch the common (pin 2) to either of the inputs Ra+ (pin 3); DEC+ (pin 4); DEC - (pin 5); and Ra- (pin 6).

All you have to do is plug in this test to the ST4 port, enter coordinate mode on the handset then press either button and watch the Ra or DEC change on the handset accordingly.

I can send you some info on building such a tester, it's cheap, just a few dollars! Also I don't know which version Gemini you have? Some versions have isolator IC's on the autoguide port, earlier version PCBs use pin 2 common as +5V! The newer PCBs use pin 2 as common or ground thru a 100ohm resistor. Indeed the issue may be in this area?...aka not allowing a signal translation correctly?

Anyway an autoguide ST4 tester will determine this!

Lastly, have you tried autoguiding with ASCOM USB (non St4)?

Get back and we will make some progress.

[kermit]

Hi Brendan,
Thank you for your help.
I have tried guiding with the ST4 port and tried 3 different cables and also with ASCOM and I get exactly the same issues.
Info on the tester would be really helpful but I not sure the problem is there as ASCOM gives me the same problem. However I think it is worth checking.
I have purchased a 12 volts 45 amps battery and tonight was the first clear night for me to test the mount powered with the battery. The battery has been fully charger and the charger indicator show that the battery is full.
When I powered the mount everything started as it should but when I came to align on the first star I got the message that the Motors Stalled. I tried a couple of time more and still got the same message.
I then powered the mount from the 12 volts transformer and everything worked fine.
Here is a graph taken in Maxim of the tracking.

https://www.flickr.com/photos/61128796@N08/15962786703/It is getting more and more confusing.
Your help is greatly appreciated.
Vince

[ZeroID]

If you think it is mains power related, ie voltage dropout or brownout then there is one guaranteed solution to fix that. Get a double conversion online UPS. Unlike smaller UPS which only intervene if they experience a mains voltage loss a double conversion UPS is always on. It takes mains input, converts to battery DC then suplies an AC output that is regulated within .3% or voltage and .1% of frequency. They are normally of the slightly bigger and more expensive variety but if you are serious about it they are the best solution. It really works because it totally isolates the load from any input variations ( including freezers and kettles etc )

We have a 3 kilowatt one running our sound system (it's expensive) and we have no more funny sub woofer crackles or buzzes and no chance of voltage spike damage or whatever. If we ever have a power outage we don't even notice, runtime is about 1:40 for our load. (TV and PS3 are also plugged into it.)
Eaton or APC are the best suppliers of this kind of gear. You might find a 1.5 Kilowatt Double Online Conversion available.

[The_bluester]

Vince, sorry in advance to sort of Hijack your thread.


Brent, it is your wife who works in the UPS sector isn't it? Could she comment on if there is a smallish double conversion unit which also has a 12VDC output direct from the battery storage? That would be the best of both worlds in my application, 240VAC uninterruptible for the AC equipment, 12VDC for the mount and heaters and battery charge state managed all in one unit.

There are two of us commonly observing and imaging respectively at the one site and if we could cover off enough load to pad both the DC and AC components of both scopes through a power failure in one box it would be worth shelling out some money on. We commonly enough have short failures, long ones would be a bit of overkill.

[wasyoungonce]

Hi Vince

The Gemini should run on +12V DC ok. The motor stall sounds like one of 2 issues:

The Battery didn't have the capacity. Being it was 45Ahr and the Gemini draws ~2.5~3A max then the battery has plenty of capacity, at least for a few hrs. or

A motor issue!

You said you had a "Motors Stalled" indication. Was this just the RA or DEC or both? If it is both, this is quite strange, which indicates the battery does not have the capacity or something is dropping the 12V!If just RA then it could be a tight worm or imbalance (yes I know you have been other this).

Back to the autoguiding.....if the autoguide jumps are with ST4 and ASCOM thru USB/RS232 then I suggest it is not an electrical issue with the main PCB (although that said, what version EPROM do you have? as there have been autoguide EPROM code fixes). It could be a motor/mechanical issue, as I have seen older motors degrade after years of use. They loose some magnetic permeability when left in stall conditions as this produces a large field current that effects the motor permanent magnet. If you have many stalls then the motor is even further degraded and looses torque.

Not that I'm saying this is the issue but it must be considered. So what I want to do is take a step back gather some basic info and then go fwd. It may seem trivial to you but this info is important so I can do a hunt around and have a think:

What mount are you using (G11?/Titian?/HGM200?/G8?/other?)?
What version Gemini PCB (http://gemini-2.com/pcb-pictures.php) do you have?
What version EPROM (http://www.docgoerlich.de/Gemini.html)is your system?
How old are you motors (guess)?
Did this issue just start one day or has it been ongoing? Was there something changed then the issue started?

Lastly...the graph does not appear that bad although DEC appears to be a little out. Do these jumps effect imaging?

Pls get back and we will take this further.

Brendan

edit:
Had a look at you Fliker images...impressive! Most appear quite round stars, so a little trailing. Thing I see is earlier images on had a little trailing and 2014 images pretty good. Also noted you using a C8?/GSO10"? on a G11? Is that correct? Have you considered mirror shift as a possible issue?
For example your image "ic-5070-curves" 2014 is an enlarged image, star roundness is good but I can see some small hot pixels arcs. Which suggests RA & DEC shift. Was this done with the C8? Mirror shift would be indicative of this...IMHO. your "m45" colour QHY10" 2014image is spot on just some colour hot pixels again showing an arc. Qhy10 Colour.


Your "M57-LRGB 2012" shows looks like a tilted image plane. The GSO images appears to show strange image tilts (collimation?), aka "best-ic6992-web, 2012".

[kermit]

Hi Brendan,
Many thanks for taking the time to help me out. It is very much appreciated.
The battery is brand sparkling new so I like to think that it is ok.
I have played with the mount following your mail and here are the finding.
1. Mount does not drive with the 12 volts battery
2. Mount works fine with the power supply. (240/13.8 volts)
3. RA motor works with battery
4. DEC motor does not work with battery but fine with power supply.
I have also this morning readjusted the worms in order to give a little bit more clearance. Maybe it will help. I clutching at straws.:)
Some history for the mount. I purchase the mount in march 2014 from a lady whose husband died prematurely. He has bought the mount for his retirement and never used it. The mount is 7 years old but only in used since I had it. When I received the mount it was in its original packaging and I could see that it was brand new. The setting of the elevation was still set for California and you have to use various spanners to allow the mount to be used in around 45 degrees north.
The mount is a Losmandy Titan 50:1
The eprom is 1.04 and I ordered today the 1.05. I don't think it will change anything but it is cheap enough.
The motors I am told by Ovision that they are the early type.
The mount never worked properly and have been sent to Ovision twice without any improvement. However Ovision is telling me that the mount is working perfectly.
Where do I get the Gemini PCB's version?
The jumps do affect the image and I am limited to a maximum of 600s exposure. After that time the image get to degraded to be useful.
At present I am using an equinox 120 and a Borg 72 for imaging and guiding so the weight is very much bellow the carrying capacity.
The image posted have been taken with a variety of instrument and the mount I used before the Titan was a Mesu 200. It was the biggest mistake of my life selling it.:(
Once again thank you for your help.
Vince

[wasyoungonce]

Yes but the battery voltage can drop rapidly under load causing the system to shutdown. For example a tight worm causes excessive motor current draw the battert voltage droops. In this case you should get a motor stall warning. This is a big giveaway, previously mentioned.



Ok so loosening the RA worm allowed it to work in RA on battery undervolting under motor load, the worm was tight. This again points to the battery and or tight worm drive mesh (since it works ok with a 13.8V PSU). Just a question....didn't you swap the RA/DEC motors...are they still in the swapped position?

The "Stalled DEC motor" indicates a bad or tight worm drive in the DEC or the motor or motor drive servo is bad. Which is why I am asking if the motors are still in the swapped position. It is usual for the RA motors to fail before the DEC was they do the majority of work!

You find your PCB version by the motor sockets and power switch. Early versions had a slide power switch and RJ (6pin telephone type) motor sockets and leads; the 2nd version had slide pwr switch and round DIN motor sockets; the final version had a rocker power switch (with internal LED) and round DIN motor sockets, all info here (http://gemini-2.com/pcb-pictures.php).

The reason I asked this as the early versions of Gemini didn't buffer the ST4 inputs and the newer ST4's were not compatible. But in truth probably all this does not matter atm as the issue appears to be either motor or worm drive.

Now, the issue goes away when using 13.8V and re-meshing allowed RA to work @12V....which points straight at the battery capacity undervolting when driving the motors and worm meshing. Most likely cause is that one or more of the RA or DEC worms were/are too tight or these is an issue with their drive...aka the oldham coupling or gearbox.

However that said, the DEC motor could be slightly bad. Here's a trick. You said the DEC would not drive @12V battery...can you just put the RA cable on the DEC motor and the DEC on the RA...leave both of these cables in their original RA/DEC sockets on the Gemini unit....so you just swap the cables on the motors...just for a test @12V battery.

What I'd ask is do this then power up and see if the DEC now works? If it does then the motor is ok and worm mesh ok...but does the RA now suffer or stall? If it does then there is a possible issue with the DEC servo drive electronics (remember the servo drive electronics are now swapped). Just do this for a short time and don't forget to put the cables back in the correct position.

If the DEC is still bad (from above test) then it is a mesh or oldham issue...mechanical. That said, @13.8v DC there is enough torque in the motor drive to overcome this "tightness"!

Then I'd suggest that the guiding jumping is caused by tight spots on the worm mesh or oldham/gearbox issues and when it reaches these spots...you get a stiction and jump.



Ahh so an ovision worm. Very nice. I don't think the EPROM is the issue here and indeed some EPROM code upgrades were specifically targeting autoguiding on the titian! That said level 1.04 has these code updates already, but this is why I asked.

The early motors??? I think Ovsion means these? See here (http://gemini-2.com/Losmandy-motors.php), around 1/2 way down the page shows the older motors...aka no cooling fins and a little shorter than the newer higher torque motors.

Lets see how this goes as the issue is certainly getting closer to resolution. Some good advice from the guys on the yahoo forum, most are much better are things like worm mesh etc than me...I'm an electronics guy and not that experienced in worm mesh matters.



Damn...don't you just hate that! That said I'm sure we can get to the bottom of it all.

However do get back as the above tests will help us narrow the field some more.

Brendan

[kermit]

Hi Brendan,
Thank you for your help. Here is the latest.
Went out this morning to buy an electric tester. Unfortunately I don't know what I am doing so I am waiting for a friend who knows how to use one of them so I can learn very quickly how to check the battery.
The PCB is the latest version with a rocker switch and the red led.
The motors are the earliest ones.
1. I have swapped the motors around and still no luck on the Dec.
2. I have swapped the Cables around still no luck.
3.I have disconnected the worm in DEC still message "motor stalled"
4. Reconnected to the Ac/dc power supply and everything is good.
I really need to check the battery to see it if is ok. Before using this time around I had it on charge and the full charged indicator what lit up when I disconnected the charger.
When I checked the battery I will let you know the score.
Many thanks
Vince

[wasyoungonce]

Thanks Vince.

I take it swapping the cables meant putting the RA on the DEC and the DEC on the RA, leaving them still plugged into the correct RA/DEC positions on the Gemini?

Which shows there is a mechanical issue on DEC.

Disconnecting the worm on DEC and it still doesn't drive is disconcerting. If it is working with the external PSU then this sounds as if mechanically it is ok (might still be a little tight in DEC worm though).

One would think even a low battery there would be enough power to turn the motor & gearbox! Motor tests by me (and motor specs) show 3.5A max current draw however, that said, this is a brushed DC motor that I have seen to start turning from as low as 1.2V with as little as 15~30~40mA (and lower).

Experience has shown that it is unusual to draw 3.5A per motor. Generally since they start turning at a lot lower current that absolute max I draw I generally in use per motor is 500~600~700mA, typicial I draw is 110mA!

One warning. A stalled motor can produce a very high current draw which can produce a large magnetic field spike that in turn can reduce the motors permanent magnet permeability. However, I'd expect this issue to only show up after many many stalls and as the motor aged!

So at this stage. The 12V battery does not have the capacity to drive the mount! But I am concerned that it couldn't even drive the motor with the worm disconnected!

Again, going back to the original issue. "jumping during autoguiding"...this appears to go hand in hand with the issues you are seeing now. A tight worm mesh will have highs and low points of contact which produce stiction which is more than likely causing these jumps.

Brendan

[kermit]

Hi Brendan.
By swapping the cables I mean on the Gemini control box I put the DEC in the RA and vice versa.
I did disconnected the worm but the DEC motor still doesn't move.
I tested the battery voltage this morning and it is 12.84 volts.
It looks like I will have to dismantle the Dec part of the mount and see it something in there is causing the problem.
Thank you for your help.
Vince

[kermit]

Hi Brendan,
Ok played with the mount today and here is the latest.
I disconnected the DEC motor from the mount so there is no stress whatsoever on the motor and you guess right it stalled.
I change everything from that point on. Cable, motor, connection on the Gemini controller and nothing would get the motor to run.
Plugged the controller on the 240/12 volts everything works.
My guess is that it is something to do with the battery but the voltage is showing 12.84 volts.
Vince

[ZeroID]

Hi Paul,
There are no UPS's that put out both mains voltage and 12 volt DC although theoretically you could tap straight into the battery as long as it didn't run a battery string which could series connect for a higher internal battery voltage.

I couldn't size up a UPS without better knowing the max load expected of all probable connected devices, Mounts, PC's, Cameras, heaters etc etc. I suspect something around the 3 Kilowatts as we have inside ( a cheap internal buy at the time ) would probably be about right. NZ$ cost on that new is about $1200. It's about the size of a good desktop PC tower. It would probably give you enough run time to at least shut down your systems properly or complete a few more images. But it certainly would protect your gear from spikes and brownouts.

Nothing to stop you plugging in a 240 VAC to 12 VDC power supply to it either although that would not necessarily be the most efficient way to solve it.

If you are using laptops you don't really need a UPS in front of those. Already isolated and protected by their PS and battery configuration.

Why not just a big marine or auto battery and a decent car charger in front of that for 12 volts ? Charger will isolate spikes and battery will provide capacity and stability. The more expensive chargers have full regulation and protection.

[BPO]

(Assuming you're using an effective multi-stage charger) -- Charge the battery. When done, disconnect the charger and connect a load to the battery, something around 10-15 Amps. A healthy battery's voltage should rapidly drop from a normal post-charge surface voltage of ~13 to ~12.5, then slowly recover to a higher voltage (~12.75) before settling down and gradually diminishing over a relatively long period to an end voltage around 11V.*

If when the load is first connected the battery immediately drops to a low voltage (~12.3 or less) and continues to drop you most likely have at least one failed cell and the battery is worthless.

Dodgy batteries often will charge up to a seemingly normal voltage but will fail under any real load. The failure rate of brand new and near new sealed batteries ("AGM, Gel, SLA" etc) is and was always far higher than that of flooded cells.

* Stop testing long before reaching this voltage... You're simply looking for any obvious faults, and there's no need to punish a healthy battery.

[The_bluester]

Thanks Brent. I split out my discussion to a new thread over here http://www.iceinspace.com.au/forum/showthread.php?t=132022 to avoid irritating the OP given my discussion is really on a different topic.

My current idea is to get a bench DC power supply able to meet my current requirements and more and use a smallish SLA or maybe even a larger car battery across the output to float me in case of outages.

I am looking at the moment into mounting the bench supply and battery in an enclosure so they can be wheeled out next to the scope and plugged in to mains via an extension cord. The battery would be separately charged before use on a proper multi stage charger to ensure it was in top condition and the bench supply would effectively maintain it.

If you have any thoughts could you reply in the other thread? I will copy this over there so that thread makes sense.

[wasyoungonce]

Now this is odd. I'd suggest the Gemini is shutting it down as the motors can turn @1.2V.

The test that putting it back on 12V PSU confirms the battery cannot run the system. Sorry you battery is bad.

[kermit]

Hi,
Ok it would seem that the battery had it. Back the shop and see what they say.
Thank you for your help.
Vince