CSF Thomson 28V 80A SMPS ex-Dolphin Network?

[Sparky67]

Hi all,

Bit of a long shot I'm sure...but I have to ask!

I have on my bench a CSF Thomson 'Alimentation 45 325 166' SMPS which I believe came from a Dolphin network base station. Although the fan runs and I can see various voltages on some of the pins the there are no volts on the main connections on the back, so I suspect some external linking is needed to make it work.

Does anyone have any info on this PSU please?

Thanks,

Martin
G4NCE

[Biggles]

Sounds like it could be a battery charger/PSU, being 28V. I did come across some PSU/chargers which actually needed to "see" a battery connected before starting up. Something to do with the low battery dropout system. Could be a similar design? I think the PSUs we had were made by Bauch. Just a thought.
Alan.

[Sparky67]

Thanks for that Alan. There are only 3 individual pins on each of the two edge connectors, the rest are blocked together, which I assume are the main outputs. I will have another look inside to see if there are perhaps steering diodes or connections to a sensing cct on any of them...

I actually stripped out a Dolphin cabin some years ago so it could be repurposed as a stand-alone TETRA base station demo site. The tech manuals were still in there. Hindsight, eh?!

Cheers,

Martin
G4NCE

[Hans PE1KWH]

Just use a 1k resistor and connect 2 of the 5 connections together with it.
Not running, next pair.... ( make a table of the possibilities ).
The 1k prevents shorts but should be low enough ( measure ) to pull up/down an input.

In some cases there are 2 connections necessary but that is mostly on the more complex server power supplies.

The worst case will be it still doesn't start but with the information from your measurings you could do an edicated guess which pin is an input and redo the connection with a lower value resistor.

[Refugee]

I would be tempted to add a pair of stacking plugs between the meter and the test leads and add the various resistors as required to get the indicated voltage within the spec of the gates and chips on the board after observing the change with a high value such as 1K.

[Biggles]

One possibility is that it is a thermal sensor input used for monitoring battery temperature. If this is the case it may be looking for a resistance value "window" to operate correctly. I'm guessing though. The Dolphin network never seemed to take off properly. It was intended as a tetra digital network for private users. I was involved with NB3, the analogue trunked national PAMR network before Dolphin and although we trialled Dolphin for a while using handportables, I went onto something else with the day job before it really got going. There must have been loads of fairly new equipment stripped out. I'm not sure how long the system was live.
Alan.

[G6Tanuki]

Also worth considering: anyone professionally designing a PSU that puts out 80A would surely have included a remote voltage-sense circuit to measure the voltage delivered to the actual load (so compensating for voltage-drop in the connecting leads) rather than the voltage at the output-terminals of the supply?

If the PSU is so designed and the 'sense' terminals are not connected to the output/load, it won't be able to to sense the delivered voltage so should go into shutdown.

[Sparky67]

Thanks Alan.

An ex-colleague tells me it is more likely to have originally been used in a .3kW or 1kW analogue TV transmitter rather than a Dolphin BS and is putting some feelers out for information. Hopefully the missing voltage mystery will soon be solved.

Hans, thanks for your suggestion. That is now plan B...

All the best,

Martin

[TonyDuell]

Originally Posted by G6Tanuki

Also worth considering: anyone professionally designing a PSU that puts out 80A would surely have included a remote voltage-sense circuit to measure the voltage delivered to the actual load (so compensating for voltage-drop in the connecting leads) rather than the voltage at the output-terminals of the supply?

If the PSU is so designed and the 'sense' terminals are not connected to the output/load, it won't be able to to sense the delivered voltage so should go into shutdown.

Most (although admittedly not all) SMPSUs with remote voltage sensing have internal resistors (100 ohms is common) between each sense terminal and the appropriate output terminal.

They are effectively shorted out by the wires to the load and sense terminals when the supply is used as intended. OK they introduce a tiny voltage error (as the sense signal now comes from a potential divider composed of the resistance of the sense wire and the internal resistor) but said resistors mean that if the sense terminal is not connected for any reason then the PSU will behave as one without remote sensing.