Please help me understand\fix this discrete linear power supply

[budm]

Re: Please help me understand\fix this discrete linear power supply

Rievax, thanks for that great handbook, so it does look like he need to have low Vf diode (page 21, fig 14) for that purpose.
Learning something everyday here. I will that in my troubleshooting guide thread.

[kaboom]

Re: Please help me understand\fix this discrete linear power supply

May be add the diode on the output of the 7805 (Anode to the output PIN OF 7805, Cathode is the out put) so that will isolate the output of 7805, the out put will be 0.6V lower but you can fix that by adding the forward bias diode and the small bypass cap in parallel between GND pin of the 7805 and GND.

Like this?

       ____
       |  |  Do
    o-----|7805|--->|---+----o V+
       |____|    |
        |      |
        +---+    |
        |  |    |
      Dg V --- C  - D430
        - ---   ^
        |  |    |
    o-------+---+-------+----o Gnd

Yes.

No. The negative rail is attempting to pull, thru the opamp, from the positive supply. A diode in series with the + output will still result in current coming out of that 7805!

You need a schottkey diode from GND to the output of that 7805; anode to GND, cathode to +5.

Instead of the backfeed coming out of the 7805, with its .7V drop, that current will now go around it, thru the schottkey. The lower junction drop of this schottkey bypass diode will take the 7805 out of this path, allowing it to start- the "fault" current is bypassed away from that 7805.

+7 is slow to come up, not because of some "mysteriously slow" transistors, but because of a bypass cap (C306, 2.2u) from +7 to the wiper of P300. It causes the regulator to back down on any sudden positive-going noise on the +7 rail.

But it also slows down the rate-of-rise when the thing's powered up with all caps discharged. When power is first applied, pin 3 on U300 is effectively tied right to the output, until C306 charges. After that, the pot and resistive divider take over to provide an accurate sample of +7.

The -7 output has a similar soft-start, but notice one thing! C307 is only 1u. And the resistors are different values, also giving a different time constant.

Increase C307 to 2.2/50, or even 4.7/50 to slow down the rate-of-rise for -7, while keeping that of +7 the same.

[budm]

Re: Please help me understand\fix this discrete linear power supply

No. The negative rail is attempting to pull, thru the opamp, from the positive supply. A diode in series with the + output will still result in current coming out of that 7805!

You need a schottkey diode from GND to the output of that 7805; anode to GND, cathode to +5.

Instead of the backfeed coming out of the 7805, with its .7V drop, that current will now go around it, thru the schottkey. The lower junction drop of this schottkey bypass diode will take the 7805 out of this path, allowing it to start- the "fault" current is bypassed away from that 7805.

+7 is slow to come up, not because of some "mysteriously slow" transistors, but because of a bypass cap (C306, 2.2u) from +7 to the wiper of P300. It causes the regulator to back down on any sudden positive-going noise on the +7 rail.

But it also slows down the rate-of-rise when the thing's powered up with all caps discharged. When power is first applied, pin 3 on U300 is effectively tied right to the output, until C306 charges. After that, the pot and resistive divider take over to provide an accurate sample of +7.

The -7 output has a similar soft-start, but notice one thing! C307 is only 1u. And the resistors are different values, also giving a different time constant.

Increase C307 to 2.2/50, or even 4.7/50 to slow down the rate-of-rise for -7, while keeping that of +7 the same.

I just read the explanation from RIEVAX Handbook that explains the function of that low Vf diode on the output.

[kaboom]

Re: Please help me understand\fix this discrete linear power supply

But the rate-of-rise of -7 can also be slowed down if need be, by increasing C307. Giving that 7805 section time to "catch up."

[budm]

Re: Please help me understand\fix this discrete linear power supply

But I still wonder why the units works before, or maybe the problem never show up during the final test before the products are released, or it is marginal so any parts that change its value by small amount caused this symptom to show up.
I am sure there must be lot of these units out there with the same problem.

[fzabkar]

Re: Please help me understand\fix this discrete linear power supply

FWIW, I notice that the parts list specifies a 1N4148 for D403 whereas Fig 4.2 in the handbook calls for a Schottky diode.

BTW the handbook was a revelation too me, too.

[budm]

Re: Please help me understand\fix this discrete linear power supply

BTW, if we do any slow down on the startup of the power supply, I.E slow down the -7V, to me that will create the problem for the 7905 too, trying to make power supply to come up at the same time is not practical, adding the diode to both 7905 and 7805 seems to be the correct approach as shown in the handbook.

[stj]

Re: Please help me understand\fix this discrete linear power supply

But I still wonder why the units works before, or maybe the problem never show up during the final test before the products are released,.

your knowlege of Sinclair is lacking,
they make a habit of selling stuff with terrible over-complex power circuits, then constantly revising them.

the Sinclair spectrum computer had 7 major revisions requiring pcb changes, and several intermediate changes with parts bridged on the pcb!

all to make THIS actually work.

[kaboom]

Re: Please help me understand\fix this discrete linear power supply

your knowlege of Sinclair is lacking,
they make a habit of selling stuff with terrible over-complex power circuits, then constantly revising them.

the Sinclair spectrum computer had 7 major revisions requiring pcb changes, and several intermediate changes with parts bridged on the pcb!

Plus, the supposed service manual for Agen24's unit didn't even show all the GNDs/commons/returns bonded/tied together- in the power supply no less! Hence his initial troubles with the "V5POS" running away.

all to make THIS actually work.

I've seen it before:


"Wheeee! Run wild, little power supply!"

[budm]

Re: Please help me understand\fix this discrete linear power supply

your knowlege of Sinclair is lacking,
they make a habit of selling stuff with terrible over-complex power circuits, then constantly revising them.

the Sinclair spectrum computer had 7 major revisions requiring pcb changes, and several intermediate changes with parts bridged on the pcb!

all to make THIS actually work.

What about Sinclair? Is this UNIT MADE BY SINCLAIR, same as Sinclair computer, excuse my ignorant! I do not see SINCLAIR name any where in the service manual, except the hand written SINCLAIR on the first page, otherwise I see the name MAURITRON.

[kaboom]

Re: Please help me understand\fix this discrete linear power supply

BTW, if we do any slow down on the startup of the power supply, I.E slow down the -7V, to me that will create the problem for the 7905 too, trying to make power supply to come up at the same time is not practical, adding the diode to both 7905 and 7805 seems to be the correct approach as shown in the handbook.

Ideally, both supplies would come up at the same time; tweaking those two softstart caps will change this... The diodes are always added as a "backup" anyway, but they must have a lower Vf than the unwanted path thru the 78xx's.

[budm]

Re: Please help me understand\fix this discrete linear power supply

"Ideally," That is the key word.

[kaboom]

Re: Please help me understand\fix this discrete linear power supply

"Ideally," That is the key word.

Those diodes are your friends...

[budm]

Re: Please help me understand\fix this discrete linear power supply

Those diodes are your friends...

Yep, after reading that handbook, thanks to RIEVAX.

[rievax_60]

Re: Please help me understand\fix this discrete linear power supply

If D403 is a 1N4148, Then replacing it with a larger silicon diode might be enough. Power diodes with low voltage ratings are supposed to have a lower forward drop.

[kaboom]

Re: Please help me understand\fix this discrete linear power supply

If D403 is a 1N4148, Then replacing it with a larger silicon diode might be enough. Power diodes with low voltage ratings are supposed to have a lower forward drop.

4148s and 400x diodes are silicon not shottkey (half-junction), with a nominal .7V drop- the same as the "backfeed drop" thru that 7805.

We must have a bypass diode whose Vf is lower than the reverse-backfeed drop of the 7805, so the diode turns on first and completely shunts the current around the 7805, allowing it to bias properly. The path thru the 7805 stays open til about .7V; a shunt path whose Vf is lower than that makes the 7805-reverse-path effectively nonexistent.

[budm]

Re: Please help me understand\fix this discrete linear power supply

"We must have a bypass diode whose Vf is lower than the reverse-backfeed drop of the 7805" So what is the level of this reverse-backfeed drop of the 7805?

[kaboom]

Re: Please help me understand\fix this discrete linear power supply

"We must have a bypass diode whose Vf is lower than the reverse-backfeed drop of the 7805" So what is the level of this reverse-backfeed drop of the 7805?

The rail in question feeds from the main +7v rail and uses a 78L05 to generate the +5vA rail. So far so good, but it doesn't always come up. When it works, it's perfectly stable at 5v, when it doesn't work, it strangely shows -0.77 volts. (I would have expected zero).

-.77V for the OP.

Current was coming up from GND, into the 78L05's GND pin, back out the output pin, from the +5 to -5 rails.

-5 came up first.

When rail-to-rail (w/o GND connections) loads were removed, that 78L05 always came up.

[budm]

Re: Please help me understand\fix this discrete linear power supply

-.77V for the OP.

Current was coming up from GND, into the 78L05's GND pin, back out the output pin, from the +5 to -5 rails.

-5 came up first.

When rail-to-rail (w/o GND connections) loads were removed, that 78L05 always came up.

Thanks, I just read more in details of the hand book chapter 2.

"There is a point to heed in applying power supply ICs. It is that a method known as "junction separation" is used
as the method of electrically separating each of the elements above. By connecting a separation region so that it is
formed by a P-type semiconductor and is the same lowest potential as the substrate, the element region and the
separation region are electrically separated and insulated by being in (PN junction) reverse bias states. If for some
reason the potential of this separation region becomes a higher potential than the element region (for example the
NPN transistor collector region in Figure 2-1), normal operation cannot be expected since the PN junction enters a
forward bias state and the separation state between the elements cannot be maintained. For example, when using a
positive output three-terminal regulator, the GND pin always must be made a lower potential than the potential of
other pins."
So the diode that I suggest to put in at the output, since the Diode to GND is clamp to 0.6V by that d403, so the added diode to the output in series as I suggested should not be on then and prevent the back feed based on this block diagram? I will do some experiment in the lab.

[rievax_60]

Re: Please help me understand\fix this discrete linear power supply

So the diode that I suggest to put in at the output, since the Diode to GND is clamp to 0.6V by that d403, so the added diode to the output in series as I suggested should not be on then and prevent the back feed based on this block diagram? I will do some experiment in the lab.

Yes, that would work if the extra drop out voltage can be spared.
Larger silicon diodes have a lower forward drop for a given current than smaller silicon diodes do.
A schottky diode is a much better option if available.