Post by thread:[EE]:Schottky rectifiers

part 1 1003 bytes content-type:text/plain; (decoded 7bit) A friend is revamping a variable output SMPS. The original used a single 24VAC winding and a 35A conventional bridge. He found that this bridge became very hot when the full 28V / 8A was being drawn and so changed to this set-up, which uses "push-pull" windings, as he calls them. He says that now there's very little heating in the diodes at full output. He wonders why Schottky diodes aren't used more often as power rectifiers. Maybe he's not looking around enough, but a quick glance through RS's catalogue shows that most if not all of their bridges aren't Schottky (I think). The other question he has is whether there's something about Schottkys he doesn't know that makes them unsuitable as power rectifiers, especially for long-term use, and should he go back (reluctantly) to standard diodes -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads part 2 1706 bytes content-type:image/gif; (decode) part 3 2 bytes -

One aspect where Schottky's are not a good as regular junction diodes is peak inverse voltage rating. I believe that Schottky's are usually rated only to about 50V or so (often less) although there are some with higher ratings, whereas run of the mill diodes with PIVs of more than 1000 volts are common. Sean At 11:18 AM 3/29/2003 +1200, you wrote: {Quote hidden}>A friend is revamping a variable output SMPS. The original used >a single 24VAC winding and a 35A conventional bridge. He found >that this bridge became very hot when the full 28V / 8A was being >drawn and so changed to this set-up, which uses "push-pull" windings, >as he calls them. He says that now there's very little heating in the >diodes at full output. He wonders why Schottky diodes aren't used >more often as power rectifiers. Maybe he's not looking around >enough, but a quick glance through RS's catalogue shows that most >if not all of their bridges aren't Schottky (I think). The other question >he has is whether there's something about Schottkys he doesn't >know that makes them unsuitable as power rectifiers, especially >for long-term use, and should he go back (reluctantly) to standard >diodes > >-- >http://www.piclist.com hint: PICList Posts must start with ONE topic: >[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads > > -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

> The other > question he has is whether there's something about Schottkys he doesn't > know that makes them unsuitable as power rectifiers, especially > for long-term use, and should he go back (reluctantly) to standard > diodes Schottky diodes can be appropriate in power supplies. The biggest impediment to more widespread use is their large reverse leakage current. Even relatively "little" 1A diodes can have 20mA reverse leakage. This tends to be highly temperature dependent, so thing may appear to work fine in the lab but fail in a less controlled environment. ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- 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

>Schottky diodes can be appropriate in power supplies. The biggest >impediment to more widespread use is their large reverse leakage current. >Even relatively "little" 1A diodes can have 20mA reverse leakage. This >tends to be highly temperature dependent, so thing may appear to work fine >in the lab but fail in a less controlled environment. Interesting... That's one of those "little" things that wasn't mentioned in a power supply basics course I took recently. Virtually every problem relating to SMPSs assumed the use of a Schottky diode, but no mention was made of reverse leakage current. --Brendan -- 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

At 11:58 AM 30/03/2003, you wrote: >>Schottky diodes can be appropriate in power supplies. The biggest >>impediment to more widespread use is their large reverse leakage current. >>Even relatively "little" 1A diodes can have 20mA reverse leakage. This >>tends to be highly temperature dependent, so thing may appear to work fine >>in the lab but fail in a less controlled environment. > >Interesting... That's one of those "little" things that wasn't mentioned in >a power supply basics course I took recently. Virtually every problem >relating to SMPSs assumed the use of a Schottky diode, but no mention was >made of reverse leakage current. SMPS's use the Schottky's for the high switching speed, not so much the lower forward voltage drop. Try using normal "slow" diodes in there and you'll find it either won't work or will heat up & die really quick! 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.

David: >>SMPS's use the Schottky's for the high switching speed, not so much >>the lower forward voltage drop. Try using normal "slow" diodes in there >>and you'll find it either won't work or will heat up & die really quick! Brendan: >Actually, that was a small portion of one of the labs. But aren't there >ultrafast diodes without the PIV and leakage limitations, but with a higher >forward drop? We use the MR852's in a switching power supply which are 200V (PIV) and 3 Amps. They have a forward drop of around 1 Volt and a reverse current of <100uA. They're much cheaper than the Schottky versions. 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.

> You're right about the diodes, and for what it's worth "fast" diode > performance varies enourmously. Recovery time is a major limitation in > higher voltage circuits with the result that there are diodes available > made from GaAs and SiC in order to combine the ultra low recovery time of > a Schottky diode with a 400v or 800v PIV. Another approach is to ensure that the diode is no longer conducting when the next pulse starts. This is not so hard to guarantee in a buck converter, but a little more tricky in a boost converter. ***************************************************************** 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.

Must (not all) of the buck designs I've seen run in continuous mode. ----- Original Message ----- From: "Olin Lathrop" <spam_OUTolin_piclistTakeThisOuTEMBEDINC.COM> To: <.....PICLISTKILLspam@spam@MITVMA.MIT.EDU> Sent: 30 March 2003 14:05 Subject: Re: [EE]:Schottky rectifiers > > You're right about the diodes, and for what it's worth "fast" diode > > performance varies enourmously. Recovery time is a major limitation in > > higher voltage circuits with the result that there are diodes available > > made from GaAs and SiC in order to combine the ultra low recovery time of {Quote hidden}> > a Schottky diode with a 400v or 800v PIV. > > Another approach is to ensure that the diode is no longer conducting when > the next pulse starts. This is not so hard to guarantee in a buck > converter, but a little more tricky in a boost converter. > > > ***************************************************************** > 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. > -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

>> Another approach is to ensure that the diode is no longer conducting >> when the next pulse starts. This is not so hard to guarantee in a buck >> converter, but a little more tricky in a boost converter. > > Must (not all) of the buck designs I've seen run in continuous mode. I don't know what you mean by "continuous mode", or how that relates to the point you responded to. Switching mode power supplies transfer individual packets of energy to from the input to the output. This means they sequence thru phases of operation. A simple buck voltage regulator, for example, has three phases. First, the pass element is enabled. This starts the output current ramping up and also charges the inductor. Second, the pass element is disabled, forcing the inductor current to come from ground thru a diode until it ramps back down to zero again. This phase is how the buck converter achieves higher output current than input current. The input current is zero (pass element off), but there is still output current. The third phase is the quiescent state. It is waiting for the output voltage to drop below the threshold to start the next pulse. The reverse recovery time of the diode is immaterial as long as a new pulse isn't started in the second phase. As I said above, this isn't hard to guarantee in a buck converter. ***************************************************************** 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.

On Sun, 30 Mar 2003, Olin Lathrop wrote: > >> Another approach is to ensure that the diode is no longer conducting > >> when the next pulse starts. This is not so hard to guarantee in a buck > >> converter, but a little more tricky in a boost converter. > > > > Must (not all) of the buck designs I've seen run in continuous mode. > > I don't know what you mean by "continuous mode", or how that relates to > the point you responded to. Switching mode power supplies transfer <snip> In Continuous mode, the current in the inductor does not drop to zero inbetween consecutive switching cycles. In discontinuous mode, the current does drop to zero. If you look at the voltage waveform between the inductor and the switch, you'll (typically) see high frequency oscillations for discontinuous mode operation. This is caused by the inductor resonating with the (relatively small) capacitance of the switch and/or catch diode. Depending on your PCB layout, this parasitic oscillation could cause excessive radiation. However, the current is typically much smaller than the current that is being switched and hence has less energy for radiating. I've seen problems emission problems with discontinuous switchers - but I've also seen others that oscillated wildly and but hardly radiated. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

{Quote hidden}> >> Another approach is to ensure that the diode is no longer conducting > >> when the next pulse starts. This is not so hard to guarantee in a buck > >> converter, but a little more tricky in a boost converter. > > > > Must (not all) of the buck designs I've seen run in continuous mode. > > I don't know what you mean by "continuous mode", or how that relates to > the point you responded to. Switching mode power supplies transfer > individual packets of energy to from the input to the output. This means > they sequence thru phases of operation. > > A simple buck voltage regulator, for example, has three phases. First, > the pass element is enabled. This starts the output current ramping up > and also charges the inductor. Second, the pass element is disabled, > forcing the inductor current to come from ground thru a diode until it > ramps back down to zero again. This phase is how the buck converter > achieves higher output current than input current. The input current is > zero (pass element off), but there is still output current. The third > phase is the quiescent state. It is waiting for the output voltage to > drop below the threshold to start the next pulse. > > The reverse recovery time of the diode is immaterial as long as a new > pulse isn't started in the second phase. As I said above, this isn't hard > to guarantee in a buck converter. Continuous mode is exactly that, the next phase starts in the second phase. Discontinuous conduction mode is what you are speaking of Olin. There are advantages to either scheme. Continuous mode is generally easier to get stable, is more linear in it's behaviour, and is MUCH easier to model. Discontinuous mode is generally harder to model and requires more filtering. However there are circumstances (such as what you describe) where discontinuous mode has it's advantages. Which mode is chosen is up to designer. Generally a converter uses either continuous or discontinuous conduction mode, very rarely will you see a converter designed for both. TTYL -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

----- Original Message ----- From: "Olin Lathrop" <olin_piclistKILLspamEMBEDINC.COM> To: <.....PICLISTKILLspam.....MITVMA.MIT.EDU> Sent: 30 March 2003 17:20 Subject: Re: [EE]:Schottky rectifiers {Quote hidden}> >> Another approach is to ensure that the diode is no longer conducting > >> when the next pulse starts. This is not so hard to guarantee in a buck > >> converter, but a little more tricky in a boost converter. > > > > Must (not all) of the buck designs I've seen run in continuous mode. > > I don't know what you mean by "continuous mode", or how that relates to > the point you responded to. Switching mode power supplies transfer > individual packets of energy to from the input to the output. This means > they sequence thru phases of operation. > > A simple buck voltage regulator, for example, has three phases. First, > the pass element is enabled. This starts the output current ramping up > and also charges the inductor. Second, the pass element is disabled, > forcing the inductor current to come from ground thru a diode until it > ramps back down to zero again. This phase is how the buck converter > achieves higher output current than input current. The input current is > zero (pass element off), but there is still output current. The third > phase is the quiescent state. It is waiting for the output voltage to > drop below the threshold to start the next pulse. > > The reverse recovery time of the diode is immaterial as long as a new > pulse isn't started in the second phase. As I said above, this isn't hard > to guarantee in a buck converter. > > > ***************************************************************** > 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. > A continuous mode converter is one that's designed to run without the quiescent part of the cycle. This confers substantial advantages at high power density as the output ripple is reduced to the point that non-electrolytic bypass capacitors are practical, the switch and diode pass near-square current pulses, the inductor is not subject to significant hysteresis loss and the control characteristic is very predictable to the point that output regulation is possible even without negative feedback. You are describing discontinuous mode. Discontinuous mode works well when combined with PFM (Pulse Frequency Modulation) in a fixed on-time and variable off time mode for minimum quiescent current. Continuous mode causes problems in boost applications because power is delivered to the inductor during transistor on-time but is only delivered to the load during off time. It can be seen that an increase in on-time will cause a decrease in delivered power in the short term, followed by an increase as the inductor current rises. This is severely detrimental to control loop stability. The loop is stabilised by ugly dominant pole compensation rather than the elegant phase lead schemes employed elsewhere. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

part 1 554 bytes content-type:text/plain; (decoded 7bit) I'm using this circuit to protect the PIC's input against voltage spikes from relays and a motor. With a 1N5818 the input is successfully stopped from going too far below 0V wrt the PIC's internal diodes. Is the 1N5818 a bit OTT and could a smaller, cheaper diode like a BAT be used ? 1N5818, 1A, Vf 0.55V BAT81, 30mA, Vf 0.40V BAT46, 150mA, Vf 0.45V BAT43/BAT86, 200mA, Vf 0.33V -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservspam_OUTTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body part 2 1332 bytes content-type:image/gif; (decode) part 3 2 bytes -

> I'm using this circuit to protect the PIC's input against voltage > spikes from relays and a motor. With a 1N5818 the input is > successfully stopped from going too far below 0V wrt the PIC's > internal diodes. Is the 1N5818 a bit OTT and could a smaller, > cheaper diode like a BAT be used ? Such wisdom is seldom seen! (And I'm serious)(and I know that could be taken as a negative comment about wisdom being lacking but it isn't meant that way :-) ). The Schottky makes an otherwise dangerous circuit safe. The smaller diodes would be fine as long as their rated currents weren't exceeded. I am certain that people will question the need for the Schottky diode there. Because of the ongoing comments from some re using body diodes for protection I recently duplicated this circuit using an actual PIC body diode in parallel with a forward conducting zener. The object was to see how much current flowed into the PIC when a forward biased zener was used for "protection". Results varied somewhat with current but AFAIR typically about 80% of the current went through the PIC! This is because the zener is not a normal silicon diode junction and has a forward diode drop substantially above that of most diodes. In the circuit shown WITHOUT the Schottky, if -12 volts was applied to the input, about 0.8 x (12-0.7)/470 = ~19 mA would flow in the PIC body diode. This would be nearly certain to disrupt PIC operation. With the Schottky diode present, ~0 mA will flow in the PIC body diode. One could ask why such a small input resistor should be used. Using a MUCH larger resistor reduces potential (no pun intended) current . 10 k= ~1 mA 100k = 0.1 mA. A proportionately smaller filter capacitor would be used if response speed was to be maintained. The 470r/10 nF gives about 5uS time constant or say 20uS response time. 100k may start to get a bit large for absolute max response with stray capacitance but it looks like you are not looking for that here. At 100k and 5 uS Tc you need ~100 F which is probably still OK. Russell McMahon -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body

> Russell McMahon wrote: > > > ...you need ~100 F which is probably still OK. > Any pointers/URL to those caps ?? > I'd like to see one... Ummm - the p seems to have vanished :-) That was, of course, 100 pF. I do have some 10F caps but that's not quite what I had in mind. However - 2700F !!! http://www.eetimes.com/news/98/991news/ultra.html 100F www-nrd.nhtsa.dot.gov/edr-site/uploads/PowerCache-PC100_capacitor.pdf 100F 300J at 2.5V, 10 x power density of typical batteries (they claim) http://www.maxwell.com/ultracapacitors/news/press_releases/2000/pc100.html Google has 60+ hits for "100 farad"! :-) RM -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email KILLspamlistservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

On Monday 31 March 2003 12:46 am, Jinx wrote: > I'm using this circuit to protect the PIC's input against voltage > spikes from relays and a motor. With a 1N5818 the input is > successfully stopped from going too far below 0V wrt the PIC's > internal diodes. Is the 1N5818 a bit OTT and could a smaller, > cheaper diode like a BAT be used ? > > 1N5818, 1A, Vf 0.55V > BAT81, 30mA, Vf 0.40V > BAT46, 150mA, Vf 0.45V > BAT43/BAT86, 200mA, Vf 0.33V Depends on the voltages and currents you're likely to be seeing. What I do is even simpler: I use a series-connected dual Schottky in a SOT-23 case (like the Fairchild BAT54S http://rocky.digikey.com/WebLib/Fairchild/Web Data/BAT54_A_C_S.pdf), running the cathode to V+, anode to GND, and common to the input pin. Then I use a resistor/capacitor to limit the current and fast pulses. The BAT54 is a 200mA 30V part (though of course the voltage rating doesn't matter here, as you're never going to see more than 5V reverse). You could use a much larger resistor than a 4.7K. If you're using this as an analog input, your external capacitor, if large enough, can provide the current to charge the sampling cap. -- Ned Konz http://bike-nomad.com GPG key ID: BEEA7EFE -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spamBeGonelistservspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body

Russell McMahon wrote: > I am certain that people will question the need for the Schottky diode > there. Because of the ongoing comments from some re using body diodes > for protection I recently duplicated this circuit using an actual PIC > body diode in parallel with a forward conducting zener. The object was > to see how much current flowed into the PIC when a forward biased > zener was used for "protection". Results varied somewhat with current > but AFAIR typically about 80% of the current went through the PIC! > This is because the zener is not a normal silicon diode junction and > has a forward diode drop substantially above that of most diodes. > > In the circuit shown WITHOUT the Schottky, if -12 volts was applied to > the input, about > > 0.8 x (12-0.7)/470 = ~19 mA > > would flow in the PIC body diode. This would be nearly certain to > disrupt PIC operation. With the Schottky diode present, ~0 mA will > flow in the PIC body diode. I was just wondering about a slight variation in the circuit. Omit the schottky diode and add a resistor (Rin2) between the zener and the PIC input pin. The difference in zener forward voltage drop and PIC body diode forward voltage drop is small, so consequently the voltage drop across the resistor is small and doesn't increase the same as the input voltage increases. For an example similar to above: Vin = 12V, Vdiode = 0.7V and Rin = 470R => Iin = 24mA. If Vpicdiode = 0.6V and Rin2 = 10k, => Inpic = 10uA. Back to the old argument of using body diode for anything, but at least keeping it in the microamp range is somewhat safer. Circuit is cheaper because the schottky is replaced with just a resistor. Input impedance to the real world is the same, so peaks are clamped the same, and input impedance to PIC is set by Rin2 and does not need to be rediculously high (where input can become suceptible to noise). Also if the zener fails (in such a fault condition where the schottky diode would blow as well) the 10k resistor (Rin2) will still provide current limiting and save the PIC. -- Brent Brown, Electronic Design Solutions 16 English Street, Hamilton, New Zealand Ph/fax: +64 7 849 0069 Mobile/txt: 025 334 069 eMail: TakeThisOuTbrent.brownEraseMEspam_OUTclear.net.nz -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body