Nap 200 ,Hicap question?.

Well I certainly didn't know an LM317 could sense the voltage output of a parallel LM317 and adjust it's output to compensate for any difference.

I'm all ears, please tell.

Via the internal output voltage sense line (which 'reads' its output voltage).

I didn't realise wikepedia went into such detail.

I'll must check it out tomorrow.

Really?

Reduced to resorting to insults now?

Nothing to learn in this post . Bye 

The internal output voltage sense line isn’t sensing the voltage output of the other regulator, it reads the output voltage of itself, enabling it to output the correct voltage. Neither regulator is trying to take control of the power rail, each regulator is simply producing an output voltage that is determined by the value of the resistors in the circuit.

Neither regulator has any interest in the voltage at the pre amp power rail.

"Neither regulator is trying to take control of the power rail, each regulator is simply producing an output voltage that is determined by the value of the resistors in the circuit."

What do you think the term 'regulator' means?  Each regulator tries to regulate (i.e. control) the voltage on it's output.

If the output of each regulator is connected to the same voltage rail, both of them will try to regulate the same voltage rail.  There must be very low resistance between each regulator and the voltage rail (this has to be the case for the regulator to do it's job), hence there is very little resistance between the outputs of the two voltage regulators.  So, as their outputs are connected together, each regulator also tries to control the output voltage of the other regulator.

The way I'm reading it, your last paragraph is contradictory to the first 2.

Surely both regulators are only controlling their own output. The voltage at the power rail is merely the combination of the two voltages.

...And therein lies the conflict!

The voltage can't simply be the combination of the two voltages - as there's very low resistance between them, the output voltage of both regulators has to be effectively the same voltage.

This evening I got my old bread board and soldering iron out, along with a couple of LM317T's and a handfull of resistors and a PC fan.

Turns out there's no fighting for control of the power rail and no dueling regulators. The regulator that outputs the lower voltage simply shuts down, presumably it detects the high voltage at it's output and does nothing other than continue monitoring, which would account for it's increase in temperature of 2 degC.

The voltage at the power rail was equal to the voltage of the higher regulator, this regulator increased in temperature by 22 degC. The same results where obtained when a single reg was used.

So, I suppose the down side of this is there's a fifty fifty chance the pre amp will be powered by the power amp and not the Hicap.

And did you test it under variable transient load conditions with capacitative loads - this is where the HF response comes into play.

I do remember that when the PMT head amp unit was connected to the newer instrumentation control amp unit (with an inbuilt head amp psu) and was also was wrongly power by the active adapter psu (designed for older equipment), then one of them would die a lot quicker than expected, and there was more HF noise in the system.  This was tracked down to strange transient HF spikes in the voltage rails from the regulators.

No I didn’t, but if I get time next weekend I may give it a go. Although I doubt very much transients and capacitors will prevent the lower output reg from shutting down.

We had to use a 60Mhz 100Ms/s DSO to catch the issue.  I don't know if were getting any effects from the propagation delays in the cabling or the wide bandwidth of the head amp that was being powered.  We also believed that it was the control circuits in the regulators that were being stressed as the devices weren't obviously overheating.