Jump to content

problem with own PSU design


Recommended Posts

Hi all,

I have a problem with my PSU. I made a prototype board, which gave good voltages.

http://home.quicknet.nl/qn/prive/alex.span/midibox/psuboardtop.jpg

So I desoldered everything, and made a new board, a little prettier.

http://home.quicknet.nl/qn/prive/alex.span/midibox/psuboardnew.jpg

The only problem now is that the 9v I measured on the old board, now measures 8v. Right before the 7809 I still measure 10,5v, this is the same as on the old board. Has the 7809 been too hot perhaps, bacause of soldering and desoldering, or do I need to seek the cause in a different direction?

Cheers,  Alex.

Link to comment
Share on other sites

Done some more testing:

I borrowed a laboratory powersupply from one of my very friendly neighbors, and tested at what voltage the 7809 actually begins to supply 9v again. That was when it's being supplied with 11,5Vdc. The only real change (beside the heating of the parts when desoldering and resoldering again) is the resistor value, at first that was 2x 1,5Kohm in series so effectively 3Kohm, and this time I used a single 1,5Kohm resistor per branch (because the serial chain solution just didn't look very pretty to me). So, for some further testing I aquired a set of 3Kohm resistors, and I will solder these in place sometime this week. I'll let you guys know what the results are.

If the 3Kohm resistors don't do the trick I'm going to replace the 7809 with a fresh one, and if that doens't do the trick either I'm going to add windings to my trafo to crank up the voltage...

Cheers,  Alex.

P.S.

If someone who understands electronics better than I do, which probably counts for everyone 'cause I'm a real n00b, feel free to comment on my actions, or to advise on actions to take!

Link to comment
Share on other sites

Hi there!

First of all: What schematic do you use? Perhaps its a failure there? To be honest, I never used a single resistor up to now when building the PSU´s by my own and I also only know PSU schematics with resistors NOT using the 78xx´s. But that again can really be me...  ;)

The 78xx should be ok, as a rule of thumb you can say the 78xx´s and 79xx´s start to work correctly with a supply voltage 2,5 volts higher than the target voltage (so 11,5 volts for a 7809 are TOTALLY ok). Also those 78xx´s are really very robust as long as you don´t mix up the pins once (one time negative voltage and that´s it - throw it away). As long as this does not happen, the 78xx´s have lots of stuff to prevent damage.

Good luck and greetz!

Link to comment
Share on other sites

Hi pay_c,

Thanks for your comments! Yeah, I suspected that I would need more volts on the Incoming pin of the 7809... here's tonight's story:

I replaced the 1,5Kohm resistors with 3Kohm resistors this evening. When I turned on the device I got a surprise, though an unpleasant one....MY FRIGGIN' BOX CAUGHT FIRE!!!  :o :o :o

One of the wires just totally melted on me, that was a pretty lightshow by the way. I immediately turned the power off and checked if anything got damaged, and fortunately every component was in the same state it was before the meltdown (I won't say they're all OK because I still don't get the voltage i've read earlier, but this could be due to the absense of a "load").

So, I replaced the cable, and tried again.... no fire this time, pffff... started measuring again... WTF!? the output voltage went further down! only 7,75 volts instead of 8!? Now I'm really pissed....

I decided this was not a good day to continue fiddeling with the 9v branch, so instead I made a nice powerLED in my box. That worked from the start, happy me!

Then I drilled some holes in the bottom of the box to neatly attach the PSU. I also made some spacers so the board is elevated from the floorboard, that all went pretty well.

To finish off the day I wanted to view my LED one more time :-)  ...still beautiful. So, I don't know why, but I took the multimeter one last time, and behold! 8,5V on the 9V branch! Still not enough, but at least better than what it has been so far with a "load" connected. Still, the 7809 is only supplied with 10 to 10,5 volts. I'll need to add windings to my trafo to solve this problem, but that type of wire is not so easy to get. I'm going to buy a second trafo (same type of course) and did as I did with this one: unwind one of the windings so I get a 8,5 / 5,7 trafo, but then I have some spare wire to make extra windings to the 8,5v side. I'll wind it up 'till it delivers about 9,2 Vac, that should be enough for the 7809 to do it's job after it's rectified.

My final schematic: psufinal.png

Cheers,  Alex.

Link to comment
Share on other sites

wow man you have to lose R1 and R2....They are probably both open now anyway if they were 1/4 watt.

Why are they there?

I would also check your filter caps, C1 and C3, as they might not have recovered so well from a load like that.  (something's gotta pay when you flame up a wire like that, I would bet it was not just the resistors.)  If you don't just swap these parts you should at least check for excessive AC ripple on your DC output lines.

Best Regards

SmashTV

Link to comment
Share on other sites

wow man you have to lose R1 and R2....They are probably both open now anyway if they were 1/4 watt.

Why are they there?

I would also check your filter caps, C1 and C3, as they might not have recovered so well from a load like that.  (something's gotta pay when you flame up a wire like that, I would bet it was not just the resistors.)  If you don't just swap these parts you should at least check for excessive AC ripple on your DC output lines.

Best Regards

SmashTV

Whoops I didnt mean for that to sound so grumpy!  ;D

Too much on my mind.....sorry about that!

Best

Smash

Link to comment
Share on other sites

I agree with smash, if r1 and r2 are supposed to be bleeder resistors then they would need to be at the output of the power supply.(worst case 13v/3000ohms=.0043A or .056W I don't think the resistors blew).  But there's some other problems.  The power transformer outputs 9 vac and 5Vac at the two secondaries.  That'll be about 12.6 Volts and 7 volts after the rectifiers. (it's been awhile so correct me someone if I'm wrong) Full wave bridge rectification is the ac x 1.414. also in your circuit as shown above, you have a 9 volt regulator supplied with a 7 volt source, and a 5 volt regulator supplied with 12 volts, switch the secondaries or the regulators.  Also the way the two supplies are built, they are separate entities, you have a ground wire connecting to both supplies via the led, pick one side or the other not both, although I don't think this would cause a problem.  More than likely you've gotten your gnds and B+ crossed somewhere and perhaps shorted the supplys' outputs or wired them in series.

My buddy Willy once told me...

"Hey go for it, Chicks dig the sparks!"

Have Fun!

RoyalScam

Link to comment
Share on other sites

wow man you have to lose R1 and R2....They are probably both open now anyway if they were 1/4 watt.

Why are they there?

I would also check your filter caps, C1 and C3, as they might not have recovered so well from a load like that.  (something's gotta pay when you flame up a wire like that, I would bet it was not just the resistors.)  If you don't just swap these parts you should at least check for excessive AC ripple on your DC output lines.

Best Regards

SmashTV

Hi Smash, thanks for your reply!

The resistors are in the schematic for some kind of protection, my first caps are rated 4700uF, my dad suggested that when using caps that big you need to provide a "load" to decipate the spiked generated by the trafo when turning the box on. I've measured the 9v circuit with and without the resistor, and it made no difference for the voltage.

The resistors are 3W and they come from his leftover military stash (must be over 40 years ago, when they still made stuff that could take a beating  ;) ), I desoldered the one in the malfunctioning branch and measured it: it's ok. I don't know how to measure/check the caps, but they are rated 40V, I don't think they've fried either. What I initially thought was that my rectifier had taken a beating as it only read 10v when being supplied with 8,5ac, but it seems that it needed a little time to recover, 'cause now everything is about fine again (thay way it used to be anyway).

I agree with smash, if r1 and r2 are supposed to be bleeder resistors then they would need to be at the output of the power supply.(worst case 13v/3000ohms=.0043A or .056W I don't think the resistors blew).  But there's some other problems.  The power transformer outputs 9 vac and 5Vac at the two secondaries.  That'll be about 12.6 Volts and 7 volts after the rectifiers. (it's been awhile so correct me someone if I'm wrong) Full wave bridge rectification is the ac x 1.414. also in your circuit as shown above, you have a 9 volt regulator supplied with a 7 volt source, and a 5 volt regulator supplied with 12 volts, switch the secondaries or the regulators.  Also the way the two supplies are built, they are separate entities, you have a ground wire connecting to both supplies via the led, pick one side or the other not both, although I don't think this would cause a problem.  More than likely you've gotten your gnds and B+ crossed somewhere and perhaps shorted the supplys' outputs or wired them in series.

My buddy Willy once told me...

"Hey go for it, Chicks dig the sparks!"

Have Fun!

RoyalScam

Oops, the schematic is wrong ofcourse, switched the 78xx's  ;D /schematic fixed now.

The ground wire needs to be there according to moebius, who made that clear during some very cryptic posts, but I measured the stuff and the wire makes no difference for both branches. The LED is ofcourse not used to connect both branches, the LED is connected between +5v and ground, did I make another mistake in my schematic? the dot should mean there's an interconnect between the 3 wires there, shouldn't it?

What I'm really curious about, is how much volts I should actually read on the 9v after rectifying, it is indeed a full wave bridge. I still believe the one rectifier might be faulty...

(The official formula would be nice too, because the AC input is not exactly 9v and 5v, rather 8,5v and 5,7v. Constantly trying to learn...)

Thanks for your help guys, it's appreciated!

Cheers,  Alex.

Link to comment
Share on other sites

If R1 and R2 are there to limit the "in rush" current to C1 and C3, then they should be in series with the capacitors instead of parallel.  Initially the capacitors have close to 0 ohms resistance (increasing to infinity as they get fully charged) and any resistance in parallel with 0 ohms is still going to be a total of 0 ohms. 

parallel: Rt = 1/(1/R1 + 1/R2)

series: Rt = R1 + R2

The internal resistance of the transformer may be enough to not need the limiting resistors.  I would just put the recommended fuse on the primary side of the transformer and only add the resistors if it blows immediately.

Link to comment
Share on other sites

Yepp, totally agree. You can (and should) get rid of those resistors. They´re not needed. And if you use a slow-blowing (german "träge") fuse, it will certainly not blow as the caps get charged the first time.

Standard PSU´s are most commonly made only with caps + recitifier, that´s just enough anyhow.

Greetz!

Link to comment
Share on other sites

Was the PSU connected to anything when you measured the 8 and 7.5 volts? You need to draw some current from most regulators before they start regulating well. According to the 7809 datasheet from Fairchild the dropout voltage for these regulators is 2V. So with a normal load attached and an input voltage of 11V everything should work.

You don't need the bleeder resistors paralel with the capacitors. The regulators will bleed the capacitors when you turn off the PSU. You could use inrush-current limiting resistors to limit current through the diode bridge. But you have to use much lower value resistors. and you have to put them in series before the diode bridge.

Link to comment
Share on other sites

Thanks for the replies!

I've spoken extensively on MSN with Mendelt, and that cleared up quite a few of my questions about the PSU. Will go and fix the thing tonight.

manvsmachine: thanks for the link! I don't really need the 5A though, so I doubt I'm going to use it. it's in my bookmarks now anyway  :)

[dutch] En ik spreek best redelijk Nederlands, dus dat is geen probleem ;) [/dutch]

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

×
×
  • Create New...