Power supply issues

[Gina]

When the heavy rain stops I'll go out to the shed and see if I can find my variable test load

Meanwhile, I've been thinking (ouch )  Some conclusions from what I've found so far :-

1. My battery charger doesn't seem to have enough power but the Fusion/Tantronics PSU does (13.8v @ 15A).
2. A large electrolytic capacitor at the scope end of the cable is advisable plus a ceramic capacitor to block HF spikes
3. Lead acid battery provides protection from over-voltage and also backup in case of mains power failure
4. If running off battery alone no additional protection would be required other than a fuse or fuses
5. If running off battery with charging at the same time a failure of the battery would remove the over-voltage protection and regulator chips would be desirable

Now I have a question.  How big an electrolytic capacitor?  I was thinking or 10,000 or 22,000 microfarad.  This obviously depends on the load current which could be 8A with separate feed for dew heaters or 10A if I run the dew heater from the same cable.  Probably worth running a separate cable for the dew heaters and no need to run them off the battery really.

[Gina]

Had a short break in the rain and went out and found the variable load.  Trouble is, it's faulty - half an amp is the max load current I can get

[Gina]

Been checking out the variable load.  Took the top off the box and checked the circuit.  Couldn't see anything wrong.  Tried it again from the 13.8v supply and it worked - 0-10A over the range of the control pot.  Put it all back together an no joy   There must be a bad connection somewhere.

[Alien 13]

Been checking out the variable load.  Took the top off the box and checked the circuit.  Couldn't see anything wrong.  Tried it again from the 13.8v supply and it worked - 0-10A over the range of the control pot.  Put it all back together an no joy   There must be a bad connection somewhere.

It might be worth checking insulation resistance if the unit has a FET mated to a heatsink with an insulating washer if its an electronic load.

Alan

[Gina]

It might be worth checking insulation resistance if the unit has a FET mated to a heatsink with an insulating washer if its an electronic load.

Alan

Thank you Alan

All transistors are the old NPN and PNP.  2N2222, 2N2905A and 4 2N2055s.  I made it a long time ago   There are far more efficient devices available these days   The 2N2055s are on two large heat sinks bolted together

Here's the circuit diagram.

[Alien 13]

Nice design should the power transistors not be 2N3055s one of my favorite devices for building HI FI power amps and in fact better in my view than any modern designed semiconductor.

PS i have a soft spot for the old OC71 still have a box full of mint mullard ones somewhere.

Alan

[Tinker1947]

Hmmm, 2N3055's rings a bell Transistor of choice for Ham Radio Power Supplies.........

[Horwig]

2N3055, that's a blast from the past, I wouldn't like to think how many of them I've blown up over the years of building power amps.

Interesting thought regarding the decoupling/smoothing capacitor at the load end. I'm going through similar thoughts, cos I keep getting bias frames that won't cancel out, and I've come to the conclusion that it could be because of psu noise.

My obsy supply is a Maplin 10a 13.8v psu, with two outputs, called clean and dirty, one being for camera and usb hub, the other for dew heaters, focuser and mount. The cable runs for both are run in close proximity from the psu to the scope, a distance of about 5m, I wonder what the breakthrough at hf is like?

As to capacitor size, depending on the stiffness of the supply, and inductance of the supply cables, a 10 to 20,000 microfarad cap would certainly clean up any crud, but something much smaller would probably suffice.

Is that answer rambly enough?

Huw

[Gina]

Nice design should the power transistors not be 2N3055s one of my favorite devices for building HI FI power amps and in fact better in my view than any modern designed semiconductor.

PS i have a soft spot for the old OC71 still have a box full of mint mullard ones somewhere.

Alan

Thank you   Sorry - yes - typo 2N3055   I well remember the old germanium OC71 - I used those a lot

Found the trouble with the circuit.  Two in fact, the intermittent fault was a dry joint on the collector of the 2N2905A.  Then the reason for total failure was that the 2N2905A had blown.  Luckily I've found another one and replaced it but even with a heat dissipator, it's getting too hot.  I think I need a TO220 style PNP transistor - the current one is the larger round type (forget the type number now) with a round finned heatsink.

[Gina]

2N3055, that's a blast from the past, I wouldn't like to think how many of them I've blown up over the years of building power amps.

Interesting thought regarding the decoupling/smoothing capacitor at the load end. I'm going through similar thoughts, cos I keep getting bias frames that won't cancel out, and I've come to the conclusion that it could be because of psu noise.

My obsy supply is a Maplin 10a 13.8v psu, with two outputs, called clean and dirty, one being for camera and usb hub, the other for dew heaters, focuser and mount. The cable runs for both are run in close proximity from the psu to the scope, a distance of about 5m, I wonder what the breakthrough at hf is like?

As to capacitor size, depending on the stiffness of the supply, and inductance of the supply cables, a 10 to 20,000 microfarad cap would certainly clean up any crud, but something much smaller would probably suffice.

Is that answer rambly enough?

Huw

Yes, absolutely ubiquitous for power supplies and power amps

I've ordered some TIP42Cs for PNP medium power transistors plus insulating spacers and washers.  I can then attach one to a heatsink - that should sort it out

I have also ordered some 4700/25v electrolytics.

[Gina]

I have the variable load working with the replacement 2N2905A but I will only use it at 10A for short periods to avoid blowing the 2N2905A.

With the battery connected by short wires to the Fusion PSU it's drawing just over 5A charging current.  Using the 10A load goes slightly over the 15A limit and the voltage drops slightly to 13.7v.  Turning off the PSU the battery gives just over 12v with a 10A load.

BTW the so called "Twin" Fusion/Tantronics PS200T is not actually a twin supply - it simply has two pairs of terminals connected together!

[Gina]

I now have the information I was seeking for the power situation

Seems I have three possible options :-

1. Run from the mains with the PS200T PSU and use 12v low dropout voltage regulators for my cameras.  (No backup facility.)
2. Run from battery and charge only with the astro gear disconnected.  (Limited running time.)
3. For longer imaging runs with backup battery for mains power failure, arrange for the supply to the astro gear to switch from PS200T to battery in the event of mains power failure.

Only option 3. covers all circumstances but is more complicated.  In the event of mains power failure, the cameras would need switching from the 12v regulated supplies to the battery line.  The mount would need the same.  The laptop controlling everything would simply switch to internal battery so no problem there.  12v 2A relays could switch the camera power and a 12v 5A relay would suffice for the mount.  The relays could be run from the PS200T.

[Gina]

Actually there is a fourth option.  Arrange that the system shuts down in the event of mains power failure, switch the mount seemlessly to battery power and park mount.  Then if the roof has been automated, close it.

[Ptarmigan]

As to capacitor size, depending on the stiffness of the supply, and inductance of the supply cables, a 10 to 20,000 microfarad cap would certainly clean up any crud, but something much smaller would probably suffice.

Is that answer rambly enough?

Yep love it !

Especially stiffness, you know your getting old when etc  LOL

Yes, I wondered if Gina really meant to use a current source symbol (without its attendant arrow) in her crowbar diag !

A big electrolytic should always be bypassed with a smaller non polarized cap, and even a pico cap if there is rf in the region. non inductive caps where reasonable.

The value is dependant upon the source impedance (including the cable) and not just the inductance. (from some of Gina's loss measurements it would seem that the resistance would be sufficient. Oooops I'll get my coat ! )

Which is where we come back to a voltage source.

I thought about a ref to pi tanks the other day Gina, but discretion etc !

Apart from removing crud there is value in such impedance/capacitance devices downstream in providing a delay to give time for crowbars to do the biz.

But crud is best first removed within the faraday cage of the psu itself, before it can get radiated or crosstalked to other cables nearby.

[kev]

Actually there is a fourth option. Arrange that the system shuts down in the event of mains power failure, switch the mount seemlessly to battery power and park mount. Then if the roof has been automated, close it.

Noooo. By doing that your stopping observing when all the local lights have gone out and its nice n dark.

I used to run my observatory off a 12v battery identical to my my car one, so if it ever gave up,,, Came in useful once.

Kev

Sent from my iPhone so excuse the typos!

[Gina]

Yep love it !

Especially stiffness, you know your getting old when etc  LOL

Yes, I wondered if Gina really meant to use a current source symbol (without its attendant arrow) in her crowbar diag !

A big electrolytic should always be bypassed with a smaller non polarized cap, and even a pico cap if there is rf in the region. non inductive caps where reasonable.

The value is dependant upon the source impedance (including the cable) and not just the inductance. (from some of Gina's loss measurements it would seem that the resistance would be sufficient. Oooops I'll get my coat ! )

Which is where we come back to a voltage source.

I thought about a ref to pi tanks the other day Gina, but discretion etc !

Apart from removing crud there is value in such impedance/capacitance devices downstream in providing a delay to give time for crowbars to do the biz.

But crud is best first removed within the faraday cage of the psu itself, before it can get radiated or crosstalked to other cables nearby.

OOoooopppsss!  You're right wrong symbol for a fuse - must see if I can create a correct one.  ATM I'm using SketchUp for circuit diagrams.  When I've got my imaging stuff sorted out and ready for use I'll set up a Linux box with XCircuit on it - best circuit diagram app I've found so far.

I will certainly bypass the electrolytic with a ceramic capacitor - probably a 100nF (I've run out of higher values - must order some more ).

The resistance is much reduced with the new cable

I would hope that the Fusion PSU would be suitably equipped with crud removal   But when they describe a single PSU as a twin I have my doubts   A ceramic capacitor across the terminals wouldn't hurt

[russellhq]

OOoooopppsss!  You're right wrong symbol for a fuse - must see if I can create a correct one.  ATM I'm using SketchUp for circuit diagrams.  When I've got my imaging stuff sorted out and ready for use I'll set up a Linux box with XCircuit on it - best circuit diagram app I've found so far.

I will certainly bypass the electrolytic with a ceramic capacitor - probably a 100nF (I've run out of higher values - must order some more ).

The resistance is much reduced with the new cable

I would hope that the Fusion PSU would be suitably equipped with crud removal   But when they describe a single PSU as a twin I have my doubts   A ceramic capacitor across the terminals wouldn't hurt

I used Fritzing http://fritzing.org/home/ when I was building my Arduino projects. It's a free open source program and worked quite well for what I needed.

[Gina]

OK fuse symbol sorted out   Also noticed a mistake in the variable load diagram - wrong transistor symbol - had an NPN where I should have had a PNP

[Gina]

Noooo. By doing that your stopping observing when all the local lights have gone out and its nice n dark.

I used to run my observatory off a 12v battery identical to my my car one, so if it ever gave up,,, Came in useful once.

Kev

Sent from my iPhone so excuse the typos!

Good point

The batteries I have for the obsy are that same size as my car battery and could also act as replacement for the car battery

[Gina]

I'm just wondering if it's safe to leave a 13.8v PSU/charger connected to a sealed lead acid battery.  I guess it probably is - car batteries are charged in the car from a voltage regulator circuit running off the alternator I believe.

EDIT - Just found this explanation by Googling which would seem to indicate that a 13.8v current limiting PSU such as the various Fusion models, are in fact, ideal backup battery chargers - keeping the battery topped up with no chance of overcharging.  A variable load can be connected and the combination will guarantee a fixed voltage to the load as the latter varies.  Then when the mains power fails the battery simply takes over.

Any comments from the experts on here are very welcome

[Gina]

Been doing a bit more checking up and it seems the Fusion PS200T does not have current limiting   However, when I tried charging my discharged battery from the PS200T the highest current I got was 9A.  Of course, it's limited by the resistance of the wires between PSU and battery.  So I think it would be safe.  In any case I will be having a fuse in the circuit (more than one in fact).  I can use a 15A (or 13A) fuse between PSU and battery and I will be having fuses for each item of equipment plus a safety fuse right at the battery (output side) to protect against a possible short circuit in the wiring.  (I'll post a circuite diagram shortly.)

This is what the Fusion PSU Instructions have to say :-

I guess I could buy a PS200ADJ but I think the PS200T I already have will do.  In fact the article I found says that 13.8v is just right for leaving the battery connected and will keep it nicely topped up to use as a backup battery supply.

[Gina]

Power circuit.  It should be obvious but the lower part with PSU, battery, fuses, power switch and digital ammeter will be in the warm room and the top part will be in a box mounted on the aluminium plate to which a couple of scopes are attached.

[Gina]

I think I have pretty much settled on using a lead acid battery topped up from the Fusion PS200T PSU as above.  Unless, of course, anyone has anything to say against this. 

There are a couple of other matters I'm giving thought to.

Firstly, whether to run the dew heaters from a separate source but then the dew heaters would not be backed up in event of a power cut.  So I guess it would make sense to run the dew heaters from the main supply.

Secondly, I'm currently running the EQ8 mount from a separate source - Maplin 13.8v 5A PSU but this will want backup as well.  I had planned on using the PS200T as two independent supplies as I thought it was a twin unit - but it isn't   With the maximum current draw for the EQ8 being 4A the Maplin PSU is quite sufficient but might be overloaded when charging the battery if the latter became fully discharged.  The Maplin PSU is rated at 9A for short periods but I'm not sure I would be happy.  I think the battery wants running either from a PSU rated at 10A continuous or with current limiting if the PSU has a lower rating.

So... replace the Maplin supply with either a beefier one or one with current limiting.  An external current limiter to place between Maplin PSU and battery would not be practical as it would need to have zero dropout.  OTOH there is one device that has zero dropout at zero amps and that's a resistor   When I had longer wires going from PSU to battery the charging current was limited to 5A when discharged.

[Gina]

I've just checked the load currents and all 4 dew heaters combined draw just 1A so I see no problem running from the main imaging system supply.  This also simplifies the wiring

The mount takes a good bit more so a separate supply for this seems best.  (With separate backup battery.)

[Gina]

So... This power circuit diagram now includes sockets for the dew heaters.