HOW does this 24V regulator put out 48V?

[jaunty]

bottom of diode bridges and transformer winding coupling

secondary is soldered in the way i found it at (grey) 15V terminal and (red) 15V terminal. The secondary makes a loop up to the fuse and back -  one side connects to the bridge rectifier at rightmost terminal (left at top of board). the other side of the secondary connects to second terminal of BR.  the the reservoir cap (yes its in the right way lol - as per markings on the board anyway) is connected to terminals 1 and 3 of the BR.

[jaunty]

view of fuse holders (end one is the blowing fuse) for the 48V rail. 4700µF/63V cap in foreground is the 48V rail reservoir cap.

[jaunty]

that's about it as far as images go.

[bdunham7]

It's still not clear, perhaps a larger all-in-one shot to get the general layout?  Could you point out the relevant components?  I assume the green and purple wires are the relevant transformer leads, but is C11 the single large black axial?  If so, I can't see the traces clearly, but is the (+) on the bridge connected to the (-) on that capacitor?  Also, I see you pointed out the heat sink for the regulator, but I don't see the connection pads? 

EDIT:  Just saw rest of photos.  It's annoying when audiophoolery stuff uses unnecessarily huge boards.

[bdunham7]

You have the wrong bridge rectifier.  The markings on the bridge need to match the function of the traces--the AC inputs are marked with a ~ and the outputs are marked + and -.  If I've compared correctly, the actual terminals of the bridge are as I've drawn them, which won't work.  And, although this bridge won't work unless you bend and swap two of the leads, you also have it in backwards.  At least that is what I'm seeing.

[jaunty]

'audiophoolery stuff' ? not at all. it's not a piece of hifi equipment. just a humble mixing board. i was not involved in the manufacture of it nor the design however. the pictures are all captioned with relevant information (i.e. how no capacitors were reversed etc etc) and which component is what.

the bridge rectifier is a BY164 and pretty much the same spec as the original B80C1500 - pinout is the same. if you reverse it (from what i understand at least) it shouldn't make a difference (or?). But i can put the original back in for you if you like. The symptoms will be the same. - ok i will try that then ... it would certainly be fantastic if this solved it but i don't really see how

[jaunty]

ok i see what you mean now. hmmm. ok i'll put the old one back in and see ...

https://www.radio741.com/en/b80c1500-1000-bridge-rectifier-fagor/23851-b80c1500-1000-bridge-rectifier-fagor.html

[jaunty]

WOW - thanks for being the 'fresh eyes' i needed ... i guess i was concentrating so much on the other regulator specs I didn't notice the pinout was all wrong ... and then all of the steps i was troubleshooting before discovering this were moot since I hadn't realized the fuse was blown (assuming the power good light meant something). So i have a very healthy 61V going into the LM7824 and 42V coming out (not sure how that works still but hey - its better than before) - numbers still aren't perfect but I feel confident i can chase down any issues now. I'm not sure if others have been through this but it's amazing sometimes what a wild goose chase we can create for ourselves (speaking only for myself here)

i owe you a bottle of something. THANK YOU!

[bdunham7]

It has the feel of audiophoolery--strange board layout, using standard components improperly, unnecessary dependence on specific or matched components such as a particular brand or even individual unit of a common part--in this case the LM7824.

You still have some issues. 

First, if you measure 61V to ground, that means that your LM7824 is seeing 43V input, but it is only rated to 40V.  That is poor design, although you might get away with it for a while.

Secondly, the reason you have 42V input makes sense at first glance because you have a 24V regulator floated up 18V, so 24+18=42.  I guess the answer is always 42.  However, the two resistors are supposed to tweak that--but resistors will only work for one particular LM7824, not all of them.  The reason such so-called standard parts will vary like this is that they have their own internal resistor set to set the voltage and while the ratios have to be the same between models, the exact values used do not.  The way to do this correctly, if this circuit can ever be called 'correct', is to replace R11 with a 5.6 to 6.3V zener diode.

If the current requirements of the circuit are low--you didn't mention the value for F4, but 'phantom power' seems to imply low power--then you could also put a 5.6V zener between the positive side of C11 and the regulator as well as a current-limiting resistor to protect against output shorts. 

[jaunty]

amazingly (maybe not amazing to you) i went through two more LM7824s but FINALLY got a stable and accurate output voltage!! so that's great. 46.5V but the spec requires ± 4.0V only. I think i found the main source of my issues - which was a bad trace that i just jumpered over.

as for the design - well who knows ... maybe it's a 'dutch thing' (made in holland). The 'audiophool' designs that i've seen before all look pretty rational to me with beefy traces and massive amounts of attention paid to star grounding etc. Usually overkill is the operative word with that stuff. This one doesn't seem terribly like that to me - but hey what would i know?

anyway - thanks again for the help! couldn't have done it without you (well possibly but i would have wasted way more time probably)

PS - forgot to answer your question - yes that fourth fuse was a 1A. it's a very low current circuit.


incidentally - service manuals for the models that came out after this all use an LM724 which is probably a much better choice - i agree that the 7824 is sketchy - at least as i understand it. either that or it's very clever. i can't imagine it was a cost saving measure as  an LM723 or LM317 would have been just as little likely

[Technomaniac]

Been trying to follow the conversation and getting befuddled.  But AS I SEE IT,  the bottom power supply uses a 24v chip, but its ground pin is returned to ground through a resistor network that lifts it well above ground, by the amount of extra voltage above 24 that's required.  The output from the transformer wasn't enough, so instead of returning the negative of the bridge rectifier to ground, it is returned to +18, which therefore provides an input to the 24v reg of whatever the bottom transformer provides, PLUS an extra 18 volts from the top supply.(That arrow that says + is pointed the wrong way. It's an INPUT from the top supply.) Presumably these two voltages add to something higher than 48 volts. The bottom regulator, because it's ground pin is not grounded, is lifted above ground by the resistors in its ground circuit, must have 24 volts on this pin.  Then the 24v regulators OUTPUT pin will instead have 24 volt PLUS whatever's on its ground pin.
First thing to do, put a lamp in place of the fuse, and measure the voltage on the regulators ground pin. It needs to be +24 volts. But remember, it has to come THROUGH the 24v reg.     But if you're blowing fuses with the reg removed, it has to be the bridge or the input electro -  I wonder if you're using fast blow fuses when you need a SLOW BLOW which will handle the  initial charging of the first electro at switchon?

[jaunty]

Thanks that was a reasonable explanation - it was kind of my understanding so far (that the 'ground' or 'common' ref for the middle pin of the LM7824 was well above zero so the regulator wasn't getting destroyed due to the 'floating' ground etc.

no fuses are blowing anymore though, luckily. (crossing fingers, knocking on wood!)- it was a stupid oversight - i'd tested various parts of the circuit (like changing to the wrong bridge rectifier) at a time when i did not realize there was a blown fuse. So it led me to believe there was no issue with the rectifier - when i found the blown fuse i replaced it and immediately started blowing fuses. With the original bridge rectifier back in i am no longer blowing fuses. I should probably look at slow blow fuses though in the future since i beefed up the reservoir caps by a good amount ...