Post by thread:[EE]: 7812 Unstable

> I am using ST 7812 regulator the problem I face is the voltage out I > get normally is constant 11.9 or 12.03 V. Some of the parts after > use for fews day or 1 month, the voltage out drop to 8.9 V and it is > unstable waving between 8.9 to 11.9 V. What is the reason of this and > how to avoid this ? > > James James, just to awake a possible problem; In 2001 we bought hundreds of 7805 produced by TI in Mexico, all of them, without exception presented several different voltages, from 4.01 to 5.12V, not a single one did show a voltage between 4.95 and 5.05V, not one. Unfortunately we only found it out, after 5 months of purchase, so the supplier "could do nothing" about it. By other technical point of view, I can suggest you to check the current and temperature over such 7812. What is the Voltage at its input? what is the load to it? Are you using any electrolytic capacitor at the 7812's output? if yes, remove it, keep only a small 100nF to filter high freq noises. Any electrolytic capacitor should be used at the input of the 78XX family, not output. The reason is that it can cause regulation delays and problems and the 78XX "can" do some messy regulation. By the same reason, the 100nF capacitor is REQUIRED to avoid high freq to mess with the regulation. Always remember, 78XX family "should not" be used when required output current is in excess of 1A, and you should always calculate the dissipated power over the 78XX. Multiply the Current by the Voltage Drop over the 78XX. It should be lower than 1W for uses without heatsink, and even using heatsink keep it lower than 4W to keep it running smooth. 15V .------. -----o---| 7812 |---o----> 12V | '--.---' | 470 _|_ | _|_ uF --- | --- 100nF | | | _|_ _|_ _|_ Vdrop over 7812 = 15-12 = 3V Suppose Current = 700mA Power = 0.7 x 3 = 2.1W It would require some small heatsink. Wagner. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

On Mon, 14 Apr 2003 13:29:47 -0400, you wrote: {Quote hidden}>> I am using ST 7812 regulator the problem I face is the voltage out I >> get normally is constant 11.9 or 12.03 V. Some of the parts after >> use for fews day or 1 month, the voltage out drop to 8.9 V and it is >> unstable waving between 8.9 to 11.9 V. What is the reason of this and >> how to avoid this ? >> >> James > > >James, just to awake a possible problem; > >In 2001 we bought hundreds of 7805 produced by TI in Mexico, all of them, >without exception presented several different voltages, from 4.01 to 5.12V, >not a single one did show a voltage between 4.95 and 5.05V, not one. > >Unfortunately we only found it out, after 5 months of purchase, so the >supplier "could do nothing" about it. > >By other technical point of view, I can suggest you to check the current >and temperature over such 7812. >What is the Voltage at its input? what is the load to it? > >Are you using any electrolytic capacitor at the 7812's output? if yes, >remove it, keep only a small 100nF to filter high freq noises. Any >electrolytic capacitor should be used at the input of the 78XX family, not >output. The reason is that it can cause regulation delays and problems and >the 78XX "can" do some messy regulation. By the same reason, the 100nF >capacitor is REQUIRED to avoid high freq to mess with the regulation. Actually, I've had 7805's go unstable with any LESS than about 10uf on the output...! -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

{Quote hidden}> >> I am using ST 7812 regulator the problem I face is the voltage out I > >> get normally is constant 11.9 or 12.03 V. Some of the parts after > >> use for fews day or 1 month, the voltage out drop to 8.9 V and it is > >> unstable waving between 8.9 to 11.9 V. What is the reason of this and > >> how to avoid this ? > >> > >> James > > > > > >James, just to awake a possible problem; > > > >In 2001 we bought hundreds of 7805 produced by TI in Mexico, all of them, > >without exception presented several different voltages, from 4.01 to 5.12V, > >not a single one did show a voltage between 4.95 and 5.05V, not one. > > > >Unfortunately we only found it out, after 5 months of purchase, so the > >supplier "could do nothing" about it. > > > >By other technical point of view, I can suggest you to check the current > >and temperature over such 7812. > >What is the Voltage at its input? what is the load to it? > > > >Are you using any electrolytic capacitor at the 7812's output? if yes, > >remove it, keep only a small 100nF to filter high freq noises. Any > >electrolytic capacitor should be used at the input of the 78XX family, not > >output. The reason is that it can cause regulation delays and problems and > >the 78XX "can" do some messy regulation. By the same reason, the 100nF > >capacitor is REQUIRED to avoid high freq to mess with the regulation. > >Actually, I've had 7805's go unstable with any LESS than about 10uf on the >output...! I usually have at least 47uF on the output of the regulator (and always a 100n) with no problems. Longish power traces on boards make this a necessity to keep things like op-amps stable when driving loads. Just don't go overboard and stick a 2200uF capacitor on the output! Also, remember to add a 100n close to the input when the reservoir cap is more the 40 or 50mm from the regulator. David... ___________________________________________ David Duffy Audio Visual Devices P/L U8, 9-11 Trade St, Cleveland 4163 Australia Ph: +61 7 38210362 Fax: +61 7 38210281 New Web: http://www.audiovisualdevices.com.au ___________________________________________ -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

> 15V .------. > -----o---| 7812 |---o----> 12V > '--.---' | > | _|_ > | --- 470 uF > | | > _|_ _|_ > > > This is the diagram I am using, the current draw out is 450 m-600 mA > only. > Any circuit improvement required to prevent output voltage oscillation ? My immediate thought is to raise the input voltage. Most voltage regulators need an input voltage *AT LEAST* 3 volts higher than the voltage they regulate, so depending upon the regulator, you may be running in a marginal condition. Check the datasheet on the regulator you're using (unless it's a low voltage dropout like the 2940 series). I'd also put a cap on the input, too, and I'd increase the size of your 470uf to something around 2200uf (or higher). -->Neil ------------------------------------------------------------------------------- Neil Bradley In the land of the blind, the one eyed man is not Synthcom Systems, Inc. king - he's a prisoner. ICQ #29402898 -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spam_OUTlistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

> This is the diagram I am using, the current draw out is 450 m-600 mA > only. > Any circuit improvement required to prevent output voltage oscillation ? Input capacitor of some sort a good idea (at least). The 470 uF is larger than usual but is probably OK. IF input goes short the 470 uF will discharge back into the regulator and MAY cause damage. If input drops rapidly from other load at turn off this may be equivalent to going short. RM {Quote hidden}> 15V .------. > -----o---| 7812 |---o----> 12V > '--.---' | > | _|_ > | --- 470 uF > | | > _|_ _|_ > > > This is the diagram I am using, the current draw out is 450 m-600 mA > only. > Any circuit improvement required to prevent output voltage oscillation ? > > James > > > {Original Message removed}

At 02:39 AM 4/16/2003 -0700, you wrote: >15V .------. >-----o---| 7812 |---o----> 12V > '--.---' | > | _|_ > | --- 470 uF > | | > _|_ _|_ > > >This is the diagram I am using, the current draw out is 450 m-600 mA >only. >Any circuit improvement required to prevent output voltage oscillation ? The output capacitor is optional for stability (it improves transient response so you should use it), but you NEED to have an input capacitor close to the regulator (from 15V to ground). Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" .....speffKILLspam@spam@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

Spehro Pefhany wrote: {Quote hidden}> At 02:39 AM 4/16/2003 -0700, you wrote: >> 15V .------. >> -----o---| 7812 |---o----> 12V >> '--.---' | >> | _|_ >> | --- 470 uF >> | | >> _|_ _|_ >> >> >> This is the diagram I am using, the current draw out is 450 m-600 mA >> only. >> Any circuit improvement required to prevent output voltage >> oscillation ? > > > The output capacitor is optional for stability (it improves transient > response so you should use it), but you NEED to have an input > capacitor close to the regulator (from 15V to ground). > > Best regards, The 78XX family can provide transient response much faster than any capacitor, its internal low impedance circuit is faster and can transport current from the primary capacitor, that is its deal, is job, is purpose. A large capacitor as this 470uF at the output is no good at all, it can damage the part and cause instability. I don't know why we still discussing about this issue, this is clearly specified by all the manufacturers, all datasheets show this issue, everyone who got stability problems with 78XX simply didn't read the instructions, when removing the output capacitor the issue was solved. Sometimes it is imperative to use output capacitors, mostly in cases where the load is able to suck large chunks of current (bigger than 5A) for very short period of time, and you will not want this current to travel via the regulator, but then, you are trading safety by instability... duh! If all your circuit current consume is lower than 1A, there is no need at all to have a large capacitor at the 78xx output, period! National 7805 PDF file says: ============ **Although no output capacitor is needed for stability, it does help transient response. (If needed, use 0.1 µF, ceramic disc). Shorting the Regulator Input: When using large capacitors at the output of these regulators, a protection diode connected input to output may be required if the input is shorted to ground. Without the protection diode, an input short will cause the input to rapidly approach ground potential, while the output remains near the initial VOUT because of the stored charge in the large output capacitor. The capacitor will then discharge through a large internal input to output diode and parasitic transistors. If the energy released by the capacitor is large enough, this diode, low current metal and the regulator will be destroyed. The fast diode will shunt most of the capacitors discharge current around the regulator. Generally no protection diode is required for values of output capacitance around 10 µF. ============ http://cache.national.com/ds/LM/LM340.pdf http://cache.national.com/ds/LM/LM7512C.pdf As a big question; Why people still not reading datasheets? Engineering at the manufacturer waste many hours to make sure all the vital information are posted at the datasheets and people just ignore it? The one who does not read the datasheet, should be the one that knows all about it, right? then why this one still with problems with a so simple issue? Wagner. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email .....listservKILLspam.....mitvma.mit.edu with SET PICList DIGEST in the body

- 3 volt *minimum* input-output difference - input either close to a large buffer cap, or add a 0.1 uF very close - output 0.1 uF (for regulator stability) close to regulator - output might also need larger cap as buffer (depends on transients in load) - don't forget the 0.1 uF at each chip that uses the stable voltage Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservspam_OUTTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

Thanks everyone for providing me ideas and solution on the previous issue. OK, the output capacitor issue is solved. The solution is not to use EL Cap at the output cap, it will cause instability. The issue now is the input capacitor to the 7812. The input capacitor I am not sure I need to add into my circuit because the input voltage is from an Unregulated Adaptor of 12 V. The distance from the Adaptor to the 7812 circuit board is 3 meter wire (10 feets wire length). The worry of adding of the input cap is if the cap short, it will burn the circuit. What is the best value to add the input cap, I get a lots of response to add around 1000 uF but I look at the 7812 datasheet, the value suppose to be 0.33 uF. If I don't add, what are the consequence and problem I might face. Thanks again. {Original Message removed}

> -----Original Message----- > From: Quah [SMTP:asiactspam_OUTTM.NET.MY] > Sent: Friday, April 25, 2003 10:14 AM > To: @spam@PICLISTKILLspamMITVMA.MIT.EDU > Subject: Re: [EE]: 7812 Unstable > > Thanks everyone for providing me ideas and solution on the previous > issue. > > OK, the output capacitor issue is solved. The solution is not to use EL > Cap at the output cap, it will cause instability. > No, that's not the issue. You can use an electrolytic cap out the output as long as you have a suitable bypass cap on the input. Because you have a long length of wire between your supply and the regulator with ascociated resiatance and inductance, you are alamost certainly to get instability if you don't decouple it properly. Why are you worried about a bypass capacitor shorting? If you spec it properly (i.e. voltage margin of around 2x) there is minimal risk of shorting. Don't use a tantalum cap if you are worried about reliability. Regards Mike ======================================================================= This e-mail is intended for the person it is addressed to only. The information contained in it may be confidential and/or protected by law. If you are not the intended recipient of this message, you must not make any use of this information, or copy or show it to any person. Please contact us immediately to tell us that you have received this e-mail, and return the original to us. Any use, forwarding, printing or copying of this message is strictly prohibited. No part of this message can be considered a request for goods or services. ======================================================================= Any questions about Bookham's E-Mail service should be directed to KILLspampostmasterKILLspambookham.com. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

> -----Original Message----- > From: Ben Jackson [SMTP:spamBeGonebenspamBeGoneBEN.COM] > Sent: Thursday, April 24, 2003 8:02 AM > To: TakeThisOuTPICLISTEraseMEspam_OUTMITVMA.MIT.EDU > Subject: Re: [EE]: 7812 Unstable > > On Fri, Apr 25, 2003 at 02:13:36AM -0700, Quah wrote: > > > > The issue now is the input capacitor to the 7812. The input capacitor I > > am not sure I need to add into my circuit because the input voltage is > > from an Unregulated Adaptor of 12 V. > > Wait, the input transformer is 12V? Doesn't the 7812 need at least > 15V input to provide a good 12V output? > A good point. An unregulated wall wart style supply gives around 16 volts off load, which droops with increasing current. This may be ok depending on the maximum current required. Mike ======================================================================= This e-mail is intended for the person it is addressed to only. The information contained in it may be confidential and/or protected by law. If you are not the intended recipient of this message, you must not make any use of this information, or copy or show it to any person. Please contact us immediately to tell us that you have received this e-mail, and return the original to us. Any use, forwarding, printing or copying of this message is strictly prohibited. No part of this message can be considered a request for goods or services. ======================================================================= Any questions about Bookham's E-Mail service should be directed to RemoveMEpostmasterTakeThisOuTbookham.com. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

> The issue now is the input capacitor to the 7812. The input capacitor I > am not sure I need to add into my circuit because the input voltage is > from an Unregulated Adaptor of 12 V. The distance from the Adaptor to > the 7812 circuit board is 3 meter wire (10 feets wire length). The > worry of adding of the input cap is if the cap short, it will burn the > circuit. Any component failing could cause, well, failures. > What is the best value to add the input cap, As big as you can afford, given price, weight, size, etc. At the very least the total output capacitance on the unregulated supply needs to be high enough to droop no less than to the minimum voltage at the maximum current. If the load current has spikes that exceed the unregulated supply's average current capacity, then that needs to be taken into account too. Also think about the 6 meters of wire between your circuit board and the power supply. That will have some resistance and inductance, therefore increasing the apparent impedence of the supply. The cap will decrease the impedence at higher frequencies. ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

15V .------. -----o---| 7812 |---o----> 12V | '--.---' | _|_ | _|_ _ _470uF| --- 104 cap | | | _|_ _|_ _|_ I found out thru' experiment that the problem of the input EL Cap of 470 uF, 25 V cause the fire spike when I insert the adaptor jack into the socket. If I use input cap of 10 uF, 50 V. The spike is smaller, still have. If I take out the input Cap from the above circuit, no more spike. Hence, I decided not to put any cap at the input of the 7812. I believe this is the better solution. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservEraseMEEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

> -----Original Message----- > From: Neil Bradley [SMTP:RemoveMEnbTakeThisOuTspamSYNTHCOM.COM] > Sent: Friday, April 25, 2003 9:47 AM > To: EraseMEPICLISTspamspamBeGoneMITVMA.MIT.EDU > Subject: Re: [EE]: 7812 Unstable > > > 15V .------. > > -----o---| 7812 |---o----> 12V > > | '--.---' | > > Unless you're using a low voltage dropout regulator, 15 volts won't be > enough to give you a stable 12 volt output. > > -->Neil > A generic 7812 has a dropout voltage of typicaly 2 volts, so it gives virtualy no margin, it should regulate at 15 volts input. Mike ======================================================================= This e-mail is intended for the person it is addressed to only. The information contained in it may be confidential and/or protected by law. If you are not the intended recipient of this message, you must not make any use of this information, or copy or show it to any person. Please contact us immediately to tell us that you have received this e-mail, and return the original to us. Any use, forwarding, printing or copying of this message is strictly prohibited. No part of this message can be considered a request for goods or services. ======================================================================= Any questions about Bookham's E-Mail service should be directed to RemoveMEpostmasterKILLspambookham.com. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservSTOPspamspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body

Actually the ones I use (National and Fairchild) claim a 2V max dropout voltage. Cheers, -Another Neil. On Friday 25 April 2003 03:47, Neil Bradley wrote: {Quote hidden}> > 15V .------. > > -----o---| 7812 |---o----> 12V > > > > | '--.---' | > > Unless you're using a low voltage dropout regulator, 15 volts won't be > enough to give you a stable 12 volt output. > > -->Neil > > --------------------------------------------------------------------------- >---- Neil Bradley In the land of the blind, the one eyed man is > not Synthcom Systems, Inc. king - he's a prisoner. > ICQ #29402898 -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spamBeGonelistservSTOPspamEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

> > > | '--.---' | > > Unless you're using a low voltage dropout regulator, 15 volts won't be > > enough to give you a stable 12 volt output. > I don't see why you'd say that. Because the data sheets for the 7812s I've used say they need a minimum 3 volts. This is quite common - not sure why it's such a surprise! > According to the datasheet the 7812 Whose 7812? There isn't a master, all telling, all knowing 7812 data sheet that covers all 7812s known to man. Like I said, it depends upon the characteristics of the specific regulator. Besides, it's never a good idea to run a part near its absolute minimum ratings. -->Neil ------------------------------------------------------------------------------- Neil Bradley In the land of the blind, the one eyed man is not Synthcom Systems, Inc. king - he's a prisoner. ICQ #29402898 -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

> > > > | '--.---' | > > > Unless you're using a low voltage dropout regulator, 15 volts won't be > > > enough to give you a stable 12 volt output. > > I don't see why you'd say that. > > Because the data sheets for the 7812s I've used say they need a minimum 3 > volts. This is quite common - not sure why it's such a surprise! National's part is speced at 2V. > > According to the datasheet the 7812 > > Whose 7812? There isn't a master, all telling, all knowing 7812 data sheet > that covers all 7812s known to man. True, at the same time you can't say that config would be unusable for ALL 7812's, it would be for some, it would be fine for others. Don't blame me for not being specific enough if you aren't specific enough. > Like I said, it depends upon the characteristics of the specific > regulator. True, we are assuming the user of the part has enough brain to look up the spec. > Besides, it's never a good idea to run a part near its absolute > minimum ratings. True, 1V margin might not be enough for some circumstances. But, define "near", again, you CAN'T make this a blanket statement, there are cases where 1V margin is more then enough. Again, if you want to make a statement make it specific. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email @spam@listserv@spam@spam_OUTmitvma.mit.edu with SET PICList DIGEST in the body

Neil, This is interesting, and very important to me, as I usually pick up 7812's based on whatever's available and low-cost. I have a ckt here feeding a 7812 from 14.2V, and can't see which mfgr the part is from. I've got 3 diff datasheets from National, Fairchild and TI, all which claim to be 2V. Can you please tell me which mfgr claims 3V? It's important to me, so that I can avoid that one. If I had to raise the input to >15V, at almost 1A, that would generate a lot more heat. Cheers, -The other Neil. On Friday 25 April 2003 11:53, Neil Bradley wrote: > ... > Because the data sheets for the 7812s I've used say they need a minimum 3 > volts. This is quite common - not sure why it's such a surprise! > ... > -->Neil -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spamBeGonelistservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

At 12:38 PM 4/25/03 -0400, Herbert Graf wrote: > > > 15V .------. > > > -----o---| 7812 |---o----> 12V > > > | '--.---' | > > > > Unless you're using a low voltage dropout regulator, 15 volts won't be > > enough to give you a stable 12 volt output. > > I don't see why you'd say that. According to the datasheet the > 7812 has a >2V dropout voltage, which means a minimum input voltage of 14V. 15V should >be fine. Unless you're talking about effects of the output impedance of the >15V source? <from your later message> > National's part is speced at 2V. You have fallen into the trap of designing around TYPICAL specs - not worst case. I'm looking at the LM340 / 78xx data sheet downloaded from National Semiconductor's site - they state quite clearly that the MINIMUM input voltage for a 7812 is 14.5 Vdc. (data sheet page 5) or 14.8 Vdc (data sheet page 6) On one hand - you are correct: a 15.0 Vdc input is entirely suitable for a 7812 regulator from NS. But please don't suggest that relying upon the TYPICAL dropout voltage of 2V for a NS 7812 is good design practice - it is not. dwayne -- Dwayne Reid <.....dwaynerspam_OUTplanet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax Celebrating 19 years of Engineering Innovation (1984 - 2003) .-. .-. .-. .-. .-. .-. .-. .-. .-. .- `-' `-' `-' `-' `-' `-' `-' `-' `-' Do NOT send unsolicited commercial email to this email address. This message neither grants consent to receive unsolicited commercial email nor is intended to solicit commercial email. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email TakeThisOuTlistserv.....TakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

At 12:33 PM 4/25/03 -0500, Picdude wrote: >Neil, > >This is interesting, and very important to me, as I usually pick up 7812's >based on whatever's available and low-cost. I have a ckt here feeding a 7812 >from 14.2V, and can't see which mfgr the part is from. I've got 3 diff >datasheets from National, Fairchild and TI, all which claim to be 2V. > >Can you please tell me which mfgr claims 3V? It's important to me, so that I >can avoid that one. If I had to raise the input to >15V, at almost 1A, that >would generate a lot more heat. I have not yet seen ANY supplier guarantee a 2V dropout for ANY 78xx regulator. As always, the data sheet remains the final authority. I quoted page numbers from the NS datasheet in my previous message. You mention Fairchild - the closest data sheet I see is for a KA78xx. Looking at page 7 of that data sheet shows that the minimum input voltage for a KA7812 is 14.5V. If your circuit can fail in a spectacular fashion because the regulator dropped out of regulation, you may wish to select a different device. However, its probably fine the way it is. How much ripple is present on the input to the regulator? Will that ripple affect your circuit if the regulator does drop out of regulation? What happens if the 12V regulated supply drops down to 11.5V? If nothing bad happens with any of the above scenarios, don't worry about it. dwayne -- Dwayne Reid <TakeThisOuTdwaynerKILLspamspamplanet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax Celebrating 19 years of Engineering Innovation (1984 - 2003) .-. .-. .-. .-. .-. .-. .-. .-. .-. .- `-' `-' `-' `-' `-' `-' `-' `-' `-' Do NOT send unsolicited commercial email to this email address. This message neither grants consent to receive unsolicited commercial email nor is intended to solicit commercial email. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email .....listservRemoveMEmitvma.mit.edu with SET PICList DIGEST in the body

> I have not yet seen ANY supplier guarantee a 2V dropout for ANY 78xx > regulator. As always, the data sheet remains the final authority. > > I quoted page numbers from the NS datasheet in my previous message. You > mention Fairchild - the closest data sheet I see is for a KA78xx. > Looking > at page 7 of that data sheet shows that the minimum input voltage for a > KA7812 is 14.5V. > > If your circuit can fail in a spectacular fashion because the regulator > dropped out of regulation, you may wish to select a different device. > > However, its probably fine the way it is. How much ripple is present on > the input to the regulator? Will that ripple affect your circuit if the > regulator does drop out of regulation? What happens if the 12V > regulated supply drops down to 11.5V? > > If nothing bad happens with any of the above scenarios, don't worry > about it. For that device in that circuit at that temperature with that load. ***************************************************************** 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 RemoveMElistservspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body

This Fairchild datasheet I'm looking at is MC7812. You can search for that at http://www.fairchildsemi.com and go to the 1st hit. I specifically pulled this datasheet up because I have used these exact regs before. Interestingly, they don't mention a 14.5V min input voltage anywhere. See page 7... But the 14.5V makes sense when you think about the o/p variation. At 12.5V o/p, 2V gives 14.5V min in. My circuit will not be affected by 12V dropping lower since at this point the only thing using the 12V are fans. And I have a measured 11.9V output (assuming my DMM is correct). But with the same DMM, the input to the 7812 is 14.2V. There is the chance that the 12V will be used by other functions, but the spec allows 12V +/- 10%. I still don't like it though. I'll raise the input to the 7812 to 14.5, which is a 1-resistor change. I spent enough time on this not to do it right in the end. And I still would like to know which chips specify 3V dropout though. If you have any clues, please let me know. Cheers, -Neil. On Friday 25 April 2003 13:43, Dwayne Reid wrote: {Quote hidden}> At 12:33 PM 4/25/03 -0500, Picdude wrote: > >Neil, > > > >This is interesting, and very important to me, as I usually pick up 7812's > >based on whatever's available and low-cost. I have a ckt here feeding a > > 7812 from 14.2V, and can't see which mfgr the part is from. I've got 3 > > diff datasheets from National, Fairchild and TI, all which claim to be > > 2V. > > > >Can you please tell me which mfgr claims 3V? It's important to me, so > > that I can avoid that one. If I had to raise the input to >15V, at > > almost 1A, that would generate a lot more heat. > > I have not yet seen ANY supplier guarantee a 2V dropout for ANY 78xx > regulator. As always, the data sheet remains the final authority. > > I quoted page numbers from the NS datasheet in my previous message. You > mention Fairchild - the closest data sheet I see is for a KA78xx. Looking > at page 7 of that data sheet shows that the minimum input voltage for a > KA7812 is 14.5V. > > If your circuit can fail in a spectacular fashion because the regulator > dropped out of regulation, you may wish to select a different device. > > However, its probably fine the way it is. How much ripple is present on > the input to the regulator? Will that ripple affect your circuit if the > regulator does drop out of regulation? What happens if the 12V regulated > supply drops down to 11.5V? > > If nothing bad happens with any of the above scenarios, don't worry about > it. > > dwayne -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spamBeGonelistserv@spam@spam_OUTmitvma.mit.edu with SET PICList DIGEST in the body

> > If nothing bad happens with any of the above scenarios, don't worry > > about it. > > For that device in that circuit at that temperature with that load. An don't forget to wave the fish the next time you want to use the circuit! If anyone wants to ask or comment on these kinds of subjects it might help to state whether the discussion is to be about either 1) something that 'can be shown to work' (implicitly: at my bench, far away from fields, at room temperature, with the propper fish waving, etc) or 2) something that should work under a wide range of conditions (temperature, fields, fishes etc) The proper answers for the two cases can be wildly differnt. Think about - unused inputs - RS-232 to PIC with only a resistor (protection diodes currents) - power supply margins (will it still work at full load and with a net voltage of - 10%?) - internal oscillators for serial asynch communication etc etc Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email TakeThisOuTlistservspammitvma.mit.edu with SET PICList DIGEST in the body

At 03:23 PM 4/25/03 -0400, Olin Lathrop wrote: > > > > If your circuit can fail in a spectacular fashion because the regulator > > dropped out of regulation, you may wish to select a different device. > > > > However, its probably fine the way it is. How much ripple is present on > > the input to the regulator? Will that ripple affect your circuit if the > > regulator does drop out of regulation? What happens if the 12V > > regulated supply drops down to 11.5V? > > > > If nothing bad happens with any of the above scenarios, don't worry > > about it. > >For that device in that circuit at that temperature with that load. Not necessarily. Read carefully what I wrote previously - I covered both situations. If the product works fine even if the regulator is in drop-out, then there is no problem. Right? Sometimes, the only need for a regulator is to ensure that the supply does not get too high. Remember: a 78xx regulator does not turn OFF if the supply is a little too low. It just passes the input voltage less the drop-out. Neil - you mention that the only use for the 12V rail is to power fans. You also mention that your 14.2V source is adjustable. Why even use the 7812 at all? Again - check the data sheet for the fans. If the allowable input supply range includes 14.2V (and it should), you may not even need the extra regulator. dwayne -- Dwayne Reid <dwaynerEraseMEplanet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax Celebrating 19 years of Engineering Innovation (1984 - 2003) .-. .-. .-. .-. .-. .-. .-. .-. .-. .- `-' `-' `-' `-' `-' `-' `-' `-' `-' Do NOT send unsolicited commercial email to this email address. This message neither grants consent to receive unsolicited commercial email nor is intended to solicit commercial email. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservEraseMEspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body

On Friday 25 April 2003 19:10, Dwayne Reid wrote: > Not necessarily. Read carefully what I wrote previously - I covered both > situations. > > If the product works fine even if the regulator is in drop-out, then there > is no problem. Right? > > Sometimes, the only need for a regulator is to ensure that the supply does > not get too high. Remember: a 78xx regulator does not turn OFF if the > supply is a little too low. It just passes the input voltage less the > drop-out. > > Neil - you mention that the only use for the 12V rail is to power > fans. You also mention that your 14.2V source is adjustable. Why even use > the 7812 at all? Again - check the data sheet for the fans. If the > allowable input supply range includes 14.2V (and it should), you may not > even need the extra regulator. Ugh! Can of worms coming up.... (but I'll try not to be wordy)... PS for a small mobo in a car. Input is 11V to 14V. No prob with the 5V, 3.3V, etc (buck-mode dc-dc converters). The 12V section was a problem -- flyback design no prob, but just couldn't get it to work, which I chalked up to a transformer prob, and decided that I was not going to learn everything I needed about xfrmr design that fast. And that was with efficiency around 75% (IIRC). But all my inductor-based converters worked ... buck, boost, etc. So I experimented with a boost to 14.2V (from the same 11-14V), which used the same chip (LM2588) and it worked well, and then I dropped that to 12V w/the 7812. Net efficiency is still 72%. The input voltage to the 7812 is adjustable by adjusting one of the feedback-voltage-divider resistors on the boost converter. Though efficiency will drop a bit. I was using the Fairchild MC7812 datasheet (still not sure if that's the chip I have), which is why I hadn't seen the 14.5V min, but it makes absolute sense with the allowable output voltage tolerance. AFAIK, the only 12V in the system is the audio, serial port circuitry and fans, and I'm not using audio or serial. The mobo allows for 12V tolerance of +/-10%, btw (ie: 10.8V to 13.2V). I've added a couple fans beyond what's on the mobo, and calculated (but haven't measured) around 800mA, which with heatsinking (the 7812), seemed to hold out okay. (But I'm thinking now that I can run the external fans right off the automotive supply and let the mobo still power the CPU fan). I may use the audio later, which is why I would like to ensure it's designed right, by raising the voltage to 14.5V. Since then, I've been able to get my hands on some COTS 1:1 transformers for the flyback design, but questioning why I should even bother. I can swap the linear 7812 for an LDO 12V reg (1.3V dropout IIRC), keep the boost at 14V, and get somewhere around 76% efficiency..... w/o rebuilding the whole PS. And I believe Natsemi has a .7V dropout unit (need to investigate) that should help raise the efficiency even more than I can get with the pesky flyback design. Cheers, -Neil. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email @spam@listservRemoveMEEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

At 07:57 PM 4/25/03 -0500, Picdude wrote: >AFAIK, the only 12V in the system is the audio, serial port circuitry and >fans, and I'm not using audio or serial. The mobo allows for 12V tolerance >of +/-10%, btw (ie: 10.8V to 13.2V). I've added a couple fans beyond what's >on the mobo, and calculated (but haven't measured) around 800mA, which with >heatsinking (the 7812), seemed to hold out okay. (But I'm thinking now that >I can run the external fans right off the automotive supply and let the mobo >still power the CPU fan). I may use the audio later, which is why I would >like to ensure it's designed right, by raising the voltage to 14.5V. This sounds like one case where you should worry about drop-out because of the audio stuff. No matter how good the PSRR on those chips is, if you have hash on the PSU rail, some of it gets into the audio. I would do just as you have suggested: move the fans directly to the raw (but filtered) auto 12V supply; raise the boost convertor output to 15V, then use a 7812 or 78L12 to get clean 12V for your audio (and serial) stuff. You should not need more than a couple hundred mA total - probably less. BTW - I'd move the CPU fan (if 12V) to the auto 12V rail as well - I find that any of those fans cause significant noise on the rail and that really can make a difference to the audio noise floor. dwayne -- Dwayne Reid <EraseMEdwayner@spam@planet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax Celebrating 19 years of Engineering Innovation (1984 - 2003) .-. .-. .-. .-. .-. .-. .-. .-. .-. .- `-' `-' `-' `-' `-' `-' `-' `-' `-' Do NOT send unsolicited commercial email to this email address. This message neither grants consent to receive unsolicited commercial email nor is intended to solicit commercial email. -- http://www.piclist.com hint: To leave the PICList @spam@piclist-unsubscribe-requestspam_OUT.....mitvma.mit.edu>

> This is interesting, and very important to me, as I usually pick up 7812's > based on whatever's available and low-cost. I have a ckt here feeding a 7812 > from 14.2V, and can't see which mfgr the part is from. I've got 3 diff > datasheets from National, Fairchild and TI, all which claim to be 2V. > > Can you please tell me which mfgr claims 3V? It's important to me, so that I > can avoid that one. If I had to raise the input to >15V, at almost 1A, that > would generate a lot more heat. I know from experience that regardless of what the data sheets say, often implementation is different than spec. More reputable manufacturers you usually can trust (TI/National), but assume that everything is going to give you problems. Don't run your designs so close to spec. I've used many 7812s from many manufacturers (TI, National, many no name brandes), some claim 3 volts as an absolute minimum, some 2 volts as you mentioned in a prior email, and even the low voltage dropout National voltage regulators. Sorry, but I'm not going to spend the time running back through data sheets to find the specific one that says it needs 3 volts minimum to work since it's a waste of time. Just because the data sheet says that it needs a minimum voltage to work doesn't mean that it's correct under all circumstances/loads, especially if it's a "low cost" 7812. If you have the ability to adjust the voltage upward, you should do so. And remember, the regulator will act differently under different current loads. But I can tell you from experience, if you're running it that near the minimum acceptable voltage rails, you're asking for trouble. If you turn it up to 16 volts, you may be burning off a bit more energy as heat, but you are saving yourself potential headaches. -->Neil ------------------------------------------------------------------------------- Neil Bradley In the land of the blind, the one eyed man is not Synthcom Systems, Inc. king - he's a prisoner. ICQ #29402898 -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

On Sunday 27 April 2003 08:32, Neil Bradley wrote: {Quote hidden}> I know from experience that regardless of what the data sheets say, often > implementation is different than spec. More reputable manufacturers you > usually can trust (TI/National), but assume that everything is going to > give you problems. Don't run your designs so close to spec. I've used > many 7812s from many manufacturers (TI, National, many no name brandes), > some claim 3 volts as an absolute minimum, some 2 volts as you mentioned > in a prior email, and even the low voltage dropout National voltage > regulators. > > Sorry, but I'm not going to spend the time running back through data > sheets to find the specific one that says it needs 3 volts minimum to work > since it's a waste of time. Just because the data sheet says that it needs > a minimum voltage to work doesn't mean that it's correct under all > circumstances/loads, especially if it's a "low cost" 7812. > > If you have the ability to adjust the voltage upward, you should do so. > And remember, the regulator will act differently under different current > loads. But I can tell you from experience, if you're running it that near > the minimum acceptable voltage rails, you're asking for trouble. If you > turn it up to 16 volts, you may be burning off a bit more energy as heat, > but you are saving yourself potential headaches. > > -->Neil Well, buffers and extra overhead is fine, but every volt I add to the input, while running at close to a full amp, adds an additional W of heat to be dissipated, so I'm trying to optimize this as much as possible. And with the 10% tolerance of the device using it, I think I've got a good buffer already. Cheers, -Neil. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

> Well, buffers and extra overhead is fine, but every volt I add to the > input, while running at close to a full amp, adds an additional W of > heat to be dissipated, so I'm trying to optimize this as much as > possible. And with the 10% tolerance of the device using it, I think > I've got a good buffer already. If you are this sensitive to the regulator overhead, either use a low dropout regulator or create your own. It won't be very hard to make a basic regulator that can do 1A at less then 2V drop with cheap parts. 2V is actually quite a bit to work with at 1A. Just don't expect the same transient response, at least without a lot of careful analisys and tweaking, but it sounds like you don't need that anyway. ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

A lot here depends on what your 15V supply is REALLY putting out. If it keeps outputting 15V even at full load, you're probably in good shape. If it's completely unregulated it probably puts out more than 15V at low loads, and less than 15V at high loads, in which case you could have problems depending on how close YOUR top load is to what the power supply thinks is full load. (OTOH, if we're talking FANS, I don't think you have much to worry about.) Have you measured what the voltage looks like at the input to the regulator at all? A simple voltmeter should be sufficient. BillW -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

On Sunday 27 April 2003 13:14, Olin Lathrop wrote: > If you are this sensitive to the regulator overhead, either use a low > dropout regulator or create your own. It won't be very hard to make a > basic regulator that can do 1A at less then 2V drop with cheap parts. 2V > is actually quite a bit to work with at 1A. Just don't expect the same > transient response, at least without a lot of careful analisys and > tweaking, but it sounds like you don't need that anyway. Exactly the plan at this point. Inevstigating a 1.3V dropout unit that I just found on NatSemi's site. THis will save me from rebuilding the whole ckt (which has the 5V, 3.3V, etc sections on it). The LDO will be a simple 3-pin-part swap. Cheers, -Neil. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

On Sunday 27 April 2003 16:14, William Chops Westfield wrote: > A lot here depends on what your 15V supply is REALLY putting out. If > it keeps outputting 15V even at full load, you're probably in good > shape. If it's completely unregulated it probably puts out more than > 15V at low loads, and less than 15V at high loads, in which case you > could have problems depending on how close YOUR top load is to what > the power supply thinks is full load. (OTOH, if we're talking FANS, > I don't think you have much to worry about.) > > Have you measured what the voltage looks like at the input to the > regulator at all? A simple voltmeter should be sufficient. > > BillW I briefly mentioned this much earlier in the thread, that I've got a measured 14.2V (very regulated) at the input to the 7812, and at full load (power-resistor to draw 1A), I get 11.9V out. Whereas I don't necessarily trust the meter is accurate to .001%, it is the same meter used on both checks. But these may/will change under different load conditions, which is why I'd like to adjust for the safety margin. Cheers, -Neil. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

>And I still would like to know which chips specify 3V dropout >though. If you have any clues, please let me know. I have a feeling that this 3V has become a "rule of thumb" when using 78xx regulators, on the basis that the data sheet gave around 2.5V. This then meant there is a bit of headroom for ripple etc. Then there is the other problem about the way you are going about things. It seems to me that you are pushing the dropout spec of the 7812, and as a result may run into strife later. However your load is "only" fans, so it is not actually affecting you if the output does actually drop out of regulation. But then again I suspect you may be able to run the fans without the regulator at all :)) -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics