fried my B40 bridge rectifier following official FM psu PDF

[vedge]

Hi,

yeah, smells burning here!

ok heres the deal, the only dual 12VAC transformer that i found was a 2.0A one

(see photo).

Is that PSU too much for my poor bridge rectifiers to handle? or i surely hooked

something wrong? 

Is there any logic for hooking up the four ~'s ? should the center cable go

to the same ~ from both bridges or it doesnt matter?

Just trying to figure out where it got wrong...

[SLP]

Hi,

four ~'s

Have a closer look at it.

there are 2 ~ , a + and a -

the 2 ~s are the ac-input and the + and the - are the dc output.

If you've connected the transformer to + and - you've caused a short-circut.

http://en.wikipedia.org/wiki/Diode_bridge

matthias

[TK.]

Hi Matthias,

yes, this was the reason. The transformer has to be connected to the AC inputs

Best Regards, Thorsten.

[vedge]

Hello

I know both ~'s of a bridge goes to AC. (and they are hooked this way)

my question is when you got two bridges for the MBFM,

and a center tap power supply,

should the same ~ of each bridge go to the center tap? or one or the

other ~ of a bridge can connect to it.

It is also possble the guy who sold it to me didnt have a clue about the pinout.

he said "longest is +, the one in diagonal to + is - and the other two are ~"

[TK.]

I think it doesn't matter, which ~ pins you select.

I made a quick diagram just to ensure:

By exchaning the ~ pins (difference between left and right circuit), from the electrical point of view nothing will change. It's like if you would flip the diodes at the vertical axis. Since the bridge is symmetrical, you would see "the same picture"

I highly recomment to add fuses to the AC lines. A short circuit at the DC side could not only blow up the bridges, but also other components!

Best Regards, Thorsten.

[vedge]

Thanks for the explanation TK.

Well, i just got back from the store with components to make another one.

(im not exaclty sure what is still alive in my old one)

Im always reclycling old 1U rack units for my projects, and all of them have fuses tied to the

AC three-prong connector. (and it blew last time)

Should i add another fuse after the tranfo too? (in the 12VAC area?)

[TK.]

Please don't add the fuse to the high-voltage lines, this is too dangerous (especially when the sockets are not isolated). Only add them to the 12V lines

Best Regards, Thorsten.

[stryd_one]

Weird... Did someone delete my reply to this?

[vedge]

Only add them to the 12V lines

So 2 fuses then?  On each  12VAC

[vedge]

argh

Second PSU, no luck... it just killed my 2A fuse i had in the 120VAC land \

is there an error on the pdf??

Ive done 3 midiboxes, one SN76489AN daughter board, which works, and another psu,

but i cant for the life of me finish this.

because of that choice of opamp, ill have to throw my MBFM project in the dumpster

is there any other opamp i can use? Dont want to be a pain, but this requirement for +-12 VDC

is beyond me. I mean i cant even hear the supposedly obvious background noise on the first sound

demos of the MBFM!

Also why on this thing:

http://www.generalguitargadgets.com/pdf/ggg_bipolar_ps.pdf

Doesnt it require a "3" cable transo? Seems any 2 cable transfo would do.

I can program efficient multithreaded modular synth plugins and hosts but i cant do a simple "real life" bipolar PSU...

you have to understand my frustration :)

[TK.]

Any PSU which delivers a bidirectional voltage will work. You could even supply the OPs with +/- 9V or +/- 15V. Have a look into your 1U racks, maybe there is already a suitable PSU which you could canibalize?

is there an error on the pdf??

Nobody else reported an error in this schematic in the past. I don't know, what should be wrong there.

Did you cross-check your circuit with the PCB of Talion? It is a nice reference!

Weird... Did someone delete my reply to this?

I hope that I haven't clicked on the delete button by fault. I remember that I did this some months ago in another posting when I moved my mouse under unstable conditions (on my sofa ;-) too quickly. But I noticed this immediately. What I want to say: if I removed your posting unintentionally, please don't take this personally!

When did you write your posting? If it was before 2007-02-11, ca. 18:00, I'm sure that it wasn't my error...

Best Regards, Thorsten.

[vedge]

Any PSU which delivers a bidirectional voltage will work. You could even supply the OPs with +/- 9V or +/- 15V. Have a look into your 1U racks, maybe there is already a suitable PSU which you could canibalize?

That was the first thing i did :)  it had +5VDC and +10VDC ... looked everywhere for a bidirectional.

Also, just checked Talion's pdf with yours, and it seems to match up.

But, back to my story. just before the fuse blown , i tried with one 12,0,12 1Amp transformer which buzzed like MAD,

but quick readouts at the end gave me proper and stable +12/-12 ... but it smelled burned,

and the plastic on the transormator melted a bit. So changed the transfo for my previous 2amp one, and that burned the 2amp fuse.

I dont know if having the digital potentiometer attached all the time at the end of the chain could have caused the overheat?

So there must be something wrongish in my board, but not enough to cause bad results at the end of the chain. I know about the pinout of the 7912, the polarity of the electrolytics... im so desperate i think im going to show my museum of horrors.

dont worry about the 2nd 2200 cap thats sideways, that was on purpose so that the thing could fit in a 1U rack space (that one was higher than the other)

Please report any mal practice, any thing i might be doing wrong. this is entirely self taught. and i accept criticism.

For a more "aesthetically pleasing" picture, try this:

http://www.midibox.org/forum/index.php?topic=8026.0

[stryd_one]

if I removed your posting unintentionally, please don't take this personally!

Hehehe not a problem, I just thought maybe I had gone a little crazy ;D

[TK.]

I don't see an error on your PCB.

Does the trouble start once the PSU is connected to the MBHP_OPL3 module, or did you measure the voltages before doing this?

The +/-12/Ground tracks on the MBHP_OPL3 module can be checked very easily against shorts if you have a multimeter or similar. Just check the resistance between the tracks, or use it as a logic tester (if this mode is available), and check for unexpected "beeps" (=short)

Best Regards, Thorsten.

[vedge]

I don't see an error on your PCB.

Thank you VERY much for looking at my pictures!

I didnt expect anyone taking the time to look at them.

Does the trouble start once the PSU is connected to the MBHP_OPL3 module, or did you measure the voltages before doing this?

i didnt dare to connect any psu to the MBHP_OLP3 yet. i want to make sure its running fine first

dont want to risk to fry the board.

As i said the voltages look fine with a multimeter at the output stage of the the psu, but the transfo either buzzes and overheats or

(the other one, just makes the 2A fuses blow)

The +/-12/Ground tracks on the MBHP_OPL3 module can be checked very easily against shorts if you have a multimeter or similar. Just check the resistance between the tracks, or use it as a logic tester (if this mode is available), and check for unexpected "beeps" (=short)

Best Regards, Thorsten.

Between      the final +12 and ground i get 35K ohm

and between the final -12  and ground i get 16K Ohm

Is that due to the different 2200 uF caps i have? (one 35v and one 50v)

Do you have explicit points and the typical resistance i should get?

Also are we sure that what i connect to my 3 input pins are the 3 cables of the "center tap 12,0,12" transfo, in that same order?

[TK.]

The resistances are ok, the intention of this test was, if you get 0 ohm (-> short) or a high resistance (-> ok)

I don't know the typical resistance, I never take care for such details, and it differs with the used parts anyhow (as you already noticed)

Also are we sure that what i connect to my 3 input pins are the 3 cables of the "center tap 12,0,12" transfo, in that same order?

The transformer connections you made are still unclear to me, but the problem is, that so many transformers look so different. I cannot really help you here, only motivate to continue with experiments.

E.g., you could doublecheck if the center cable is really the center cable by measuring the voltages between the 3 transformer outputs. The voltage difference between this cable and the outer should be 12V, the voltage difference between the outer cables should be 24V

Best Regards, Thorsten.

[vedge]

IT WORKS!

more explanations later but i changed the design of my PSU, and i was able to recycle the two psus

i made.

Now .. too much fun playing with it...

well Ok one channel doesn't work, but that doesnt stop me from having fun right now ;)

[TK.]

Great!!! :)

(because I was really out of ideas ;-)

Best Regards, Thorsten.

[vedge]

Its that dual-Rectifier thing.

A friend of mine (attending college in electronics), to whom i've shown the suggested MBFM PSU schematics

told me that this was a recipe for disaster, so i removed one of the rectifier and rewired the input stage of my boards a bit like this:

http://www.tonmeister.ca/main/textbook/img600.png

(like his suggestions, but this picture has a typo, shows both +12, when the lower one should be -12)

Then both my psus's worked fine. No rattling transformers or exploding fuses.

Sadly he didn't take the time to explain why it was dangerous, i remember something vague about

"jumping negatives" or "jumping grounds" or something like that, its not my field so ill let others argue.

In any case. Thanks for the help, i now have a nice MBFM that i can play with.

[TK.]

Now, this is interesting!

I should consider to overwork the schematic

Best Regards, Thorsten.

[TheAncientOne]

There is a problem with the diagram.

If you look at the right hand drawing, the middle two diodes in the left hand side of the  are effectively back to back, thus connecting the centre tap to the ground, with a drop of about 0.7V, AC.

Worse, if you look at the right hand pair, when the lower winding is positive, (and thus the upper one negative), they are in series across the windings, effectively shorting them out!

If you have a centre tapped secondary, use 1 bridge, AC1 and AC2 to the outer leads, and connect the centre tap to ground.

If you have two seperate secondaries, (or two seperate tramsformers), connect a bridge to each, and then common a positive and a negative to ground.

HELP! what is a decent free or cheap program I can use to jot down quick circuit diagrams ? I've had a go in eagle, but I need more practice...

Hope this helps

Mike

[vedge]

If you look at the right hand drawing

Which one are you refering to:

This: http://www.ucapps.de/mbhp/mbhp_opl3_psu.pdf

Or

That: http://www.generalguitargadgets.com/pdf/ggg_bipolar_ps.pdf

[TheAncientOne]

Which one are you refering to:

This: http://www.ucapps.de/mbhp/mbhp_opl3_psu.pdf

Or

That: http://www.generalguitargadgets.com/pdf/ggg_bipolar_ps.pdf

I was referring to the pictures in TK's post. The http://www.ucapps.de/mbhp/mbhp_opl3_psu.pdf

is the one I think is wrong. If you follow the diode path in TK's right hand drawing you'll see what I mean.

I'm working away this weekend, when I get home I'll proto it up and check to make sure.

This is a good circuit:http://www.musicfromouterspace.com/analogsynth/powersupply.html

Basic thing is: you only need one bridge!

Mike

[TheAncientOne]

Did some more checking this weekend.

The end result is this:

Use 2 bridge rectifiers with two seperate secondaries

Use one bridge with a centre tapped secondary.

The diagrams at:

http://www.tonmeister.ca/main/textbook/img600.png

and

http://www.musicfromouterspace.com/analogsynth/powersupply.html

Are both good.

The schematic at

http://www.generalguitargadgets.com/pdf/ggg_bipolar_ps.pdf

is OK if you only have a single secondary winding and need plus and minus voltages. It won't suply as much current as a centre tapped secondary, because it's only half wave rectifying the AC for each regulator.

Once I can get up to speed on something that will let me make decent schematics for the web, I'll try and do a  proper write up.

Mike

[TK.]

Thank you very much for the research!

I've updated the OPL3 PSU schematic, so that it only uses a single bridge rectifier. I think that I never found this issue by myself, since I'm using transformers with seperate secondaries - and I'm not sure, if I connected the two middle AC lines together. I will check this once I open the case of my MBFM again.

The PSU of JD Sleep looks like a good alternative solution for people, who have a single secondary transformer lying around which they want to re-use for the AOUT or OPL3 module.

A wiki page about this topic would be great! I would add a link to this page from ucapps.de, so that everybody can read more about different variations.

Freeware schematic drawing tool: I'm using xcircuit since more than 10 years now. I don't know if there are better tools in the meantime, but this one works under different operating systems, which is a big advantage.

Best Regards, Thorsten.