Post by thread:[PIC]: Need help with a stupid problem

> Attached file as you can see its a small and simple psu. > this device produce heat to much, it only supply to a opamp about 10-15mA . > the regulators are very hot as well the input capacitors. I use large caps' > 2200uf each (not 1000uf as in the schematic. could it be the reason for the > heating? > just let me mention that it's a phono pre-amp so I use so large caps. > furthermore the transformer is 220v-2x18vac What is the voltage rating of the elco's? For 18 V AC in you would need (at least) 18 x sqrt(2) x 1.2 = 30.5 Volt. Wouter van Ooijen -- Van Ooijen Technische Informatica: http://www.voti.nl Jal compiler, Wisp programmer, WLoader bootloader, PICs kopen -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spam_OUTlistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

Tal Bejerano - AMC wrote: > Hi 2 All > > Attached file as you can see its a small and simple psu. > this device produce heat to much, it only supply to a opamp about 10-15mA . > the regulators are very hot as well the input capacitors. I use large caps' > 2200uf each (not 1000uf as in the schematic. could it be the reason for the > heating? > just let me mention that it's a phono pre-amp so I use so large caps. > furthermore the transformer is 220v-2x18vac It looks as though you're dropping a lot more voltage in those regulators than you need to. A 2x12V transformer would probably be a better choice for those +- 9V outputs. As someone else already mentioned, you also have to make sure that your capacitor voltage ratings are suitable. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email .....listservKILLspam@spam@mitvma.mit.edu with SET PICList DIGEST in the body

Sigh! This sort of issue comes up again and again. Bypass Bypass Bypass and read the data sheet which says bypass as close as possible to the leads, especially if large filter capacitors are used. Put in a couple 0.01 or 0.1 uF caps right on the leads of the regulators. And like the rest of the postings on this subject, I'm assuming you have used at least 35V capacitors here (1.414* 18V). For low noise in linear amplifiers there is a better way too. Use a 7812 and 7912 regulator to create 12V and regulate most of your power supply to a usable value. Most of the heat will be dissipated over these devices. Then use a second pair of the 7809,7909 regulators to create your analogue voltages from the +/-12V And bypass bypass bypass. John Dammeyer Wireless CAN with the CANRF module. www.autoartisans.com/documents/canrf_prod_announcement.pdf Automation Artisans Inc. Ph. 1 250 544 4950 > {Original Message removed}

If you're stuck with that transformer, just use one secondary and ground one leg. The other leg to the anode and cathode of each diode feeding each capacitor respecively in a half wave configuration. Since you are not using much current, you can get away with it. If the cap voltage ratings were sufficient, the caps are probably getting hot from the natural conduction of heat from the regulators since you said it was a tight close assembly. Rick Tal Bejerano - AMC wrote: {Quote hidden}> Hi 2 All > > Attached file as you can see its a small and simple psu. > this device produce heat to much, it only supply to a opamp about 10-15mA . > the regulators are very hot as well the input capacitors. I use large caps' > 2200uf each (not 1000uf as in the schematic. could it be the reason for the > heating? > just let me mention that it's a phono pre-amp so I use so large caps. > furthermore the transformer is 220v-2x18vac > > Regards > > Tal Bejerano > AMC - ISRAEL > > ------------------------------------------------------------------------ > Name: phono psu.jpg > phono psu.jpg Type: JPEG Image (image/jpeg) > Encoding: base64 -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

> Attached file as you can see its a small and simple psu. > this device produce heat to much, it only supply to a opamp about 10-15mA . > the regulators are very hot as well the input capacitors. You are trying to make a +- 9V regulated supply. The transformer secondaries are too high voltage for this, which will cause unnecessary heating in the linear regulators. The unregulated voltages will be 18 * sqrt(2) = 25V, possibly higher depending on input line voltage and what load the 18V is specified at. 25 - 9 = 16 volts drop on each regulator. 16V * 15mA = 240mW. That should cause noticeable heating in a TO-5 package, but should still be within tolerance. It should certainly not be "very hot", so something else is going on. The capacitors shouldn't be dissipating any power. They are either being heated indirectly by the regulators, or something is very seriously wrong. Are you sure the capacitors are installed the right way? If not they will draw excessive current, get hot, and explode in a minute or so. I'm serious about this. Don't put your face near this circuit if there is a chance the capacitors are installed backwards. They are drawn correctly on the schematic. Are you sure the load only draws 15mA? Is that what it's supposed to draw or what you've actually measured? ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email .....listservKILLspam.....mitvma.mit.edu with SET PICList DIGEST in the body

> If you're stuck with that transformer, just use one secondary and ground one > leg. The other leg to the anode and cathode of each diode feeding each > capacitor respecively in a half wave configuration. Since you are not using > much current, you can get away with it. If the cap voltage ratings were > sufficient, the caps are probably getting hot from the natural conduction of > heat from the regulators since you said it was a tight close assembly. Um Rick, the peak voltage will still be the same. It will only happen once every power cycle instead of twice. Since he's drawing so little current (supposedly) a half wave rectifier should suffice, but it will produce the same (actually slightly higher due to one less diode drop) unregulated voltage. If you've got the extra transformer secondary, you might as well use it. ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservspam_OUTTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

Olin About the caps' I found that I use 16v instead of using 35v. the transformer is what I have in my junk box, you know those who are packet in plastic case the deliver about 40mA. I will buy 12vx2 tomorrow and check again. about the consumption: I didn't measure it but data say "6mA" it's a LM4558. other problem I have is "hum" look like main 50hz. I feel handicapped with out a scope but I try to reduces it in many ways. it's really a bad project for cause this IC is very sensitive. Regards Tal Bejerano AMC - ISRAEL {Original Message removed}

> About the caps' I found that I use 16v instead of using 35v. > the transformer is what I have in my junk box, you know those who are packet > in plastic case the deliver about 40mA. I will buy 12vx2 tomorrow and check > again. > about the consumption: I didn't measure it but data say "6mA" it's a LM4558. Does that include the load? Probably not. > other problem I have is "hum" look like main 50hz. If your caps were getting warm, then they were drawing current, which would aggrevate the hum. I wouldn't chase that until the caps are right and the voltages look reasonable with no load. Then connect a known load, like a resistor, and see how much the supply droops and if anything gets hot. ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email @spam@listservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

It's very common for hum to sneak into simple analog circuits from not-so-obvious places. Here are a couple of things to look for: 1. The layout of your actual power supply wiring. Make absolutely sure that you create a central "magic point" at which the transformer center-tap wire and the two 1000uF electrolytic capacitors all join together. Then run a bus from exactly this point out to the regulators, and the to rest of your circuit. This forces all the very nasty charging currents to keep to themselves, rather than coupling into your amplifier. 2. You mention that you're using a 4558 opamp. In no way should this be a "sensitive" part, it's very tame and well-behaved. Again, local power supply bypassing is critical. Be sure there's something like a 0.1uF cap from each of the opamp's power terminals to the nearest circuit ground point, and that these caps are physically very near the opamp. Without these, your circuit may very well have a high-frequency parasitic oscillation, and -- surprise -- the resulting common-mode conduction in the opamp's output stage can draw an unexpectedly large (but still not fatal) current and things will overheat. Audible hum is a common manifestation of such an oscillation, because the opamp is operating in a sort-of unintended mode and its power supply rejection suffers. BTW, there's nothing at all "stupid" about the problems you're experiencing. We've all been there. These oh-so-simple circuits can misbehave in an astounding number of ways, you just have to keep at it till you've learned the right tricks. ;-) ----- Original Message ----- From: "Tal Bejerano - AMC" <KILLspamkooterKILLspamZAHAV.NET.IL> To: <RemoveMEPICLISTTakeThisOuTMITVMA.MIT.EDU> Sent: Thursday, May 30, 2002 3:45 PM Subject: Re: [PIC]: Need help with a stupid problem > Olin > > About the caps' I found that I use 16v instead of using 35v. > the transformer is what I have in my junk box, you know those who are packet > in plastic case the deliver about 40mA. I will buy 12vx2 tomorrow and check > again. > about the consumption: I didn't measure it but data say "6mA" it's a LM4558. > other problem I have is "hum" look like main 50hz. I feel handicapped with > out a scope but I try to reduces it in many ways. it's really a bad project > for cause this IC is very sensitive. > > > Regards > > Tal Bejerano > AMC - ISRAEL > > > {Original Message removed}

You toasted your caps. You'll need new ones even at 12Vx2 transformer: I'd recommend at least 20V preferably 25V. Bob Ammerman ----- Original Message ----- From: "Tal Bejerano - AMC" <spamBeGonekooterspamBeGoneZAHAV.NET.IL> To: <TakeThisOuTPICLISTEraseMEspam_OUTMITVMA.MIT.EDU> Sent: Thursday, May 30, 2002 6:45 PM Subject: Re: [PIC]: Need help with a stupid problem > Olin > > About the caps' I found that I use 16v instead of using 35v. > the transformer is what I have in my junk box, you know those who are packet > in plastic case the deliver about 40mA. I will buy 12vx2 tomorrow and check > again. > about the consumption: I didn't measure it but data say "6mA" it's a LM4558. > other problem I have is "hum" look like main 50hz. I feel handicapped with > out a scope but I try to reduces it in many ways. it's really a bad project > for cause this IC is very sensitive. > > > Regards > > Tal Bejerano > AMC - ISRAEL > > > {Original Message removed}