Matronics Email Lists :: View topic

Matronics Email Lists
Web Forum Interface to the Matronics Email Lists

Get Email Distribution Too! FAQ Search Memberlist Usergroups Register

Profile Log in to check your private messages Log in

Bridge Diodes use
Goto page 1, 2 Next

Matronics Email Lists Forum Index -> AeroElectric-List

View previous topic :: View next topic

Author Message

jskiba(at)icosa.net
Guest

Posted: Wed May 28, 2008 9:11 pm Post subject: Bridge Diodes use

Okay, if you use the bride diodes to power a device from two power sources.



Which one will carry the load?

Or will they both share the load in some way?



I thought I read or heard someplace that the power source with the higher

voltage takes the load but how much? All of it?



I hope this question makes sense.....



For example if a device draws 10 amps and is powered by bridge diode with

power source 1 at 13.8 volts and power source 2 at 13 volts



Will source 1 take the whole load all 10 amps ?

Or will they share it in some fashion and how do you calculate that ?



I ask since I want to make sure I do not over load my two power source

config.



Thanks

Jeff.

nuckolls.bob(at)cox.net
Guest

Posted: Thu May 29, 2008 2:32 am Post subject: Bridge Diodes use

At 12:05 AM 5/29/2008 -0500, you wrote:

 	  Quote: 	
		  



Okay, if you use the bride diodes to power a device from two power sources.



Which one will carry the load?

Or will they both share the load in some way?


	




 	  Quote: 	
		  I thought I read or heard someplace that the power source with the higher

voltage takes the load but how much? All of it?


	


    correct

 	  Quote: 	
		  I hope this question makes sense.....



For example if a device draws 10 amps and is powered by bridge diode with

power source 1 at 13.8 volts and power source 2 at 13 volts



Will source 1 take the whole load all 10 amps ?

Or will they share it in some fashion and how do you calculate that ?



I ask since I want to make sure I do not over load my two power source

config.


	


   Diodes used in this manner are not intended to

   be load sharing devices. Their only function is

   to make sure the load has voltage from at least

   one source AND to prevent backfeed of energy from

   a good source into a failed source.



   Thus they are electrical ISOLATION.



   Bob . . .

        ----------------------------------------)

        ( . . .  a long habit of not thinking   )

        ( a thing wrong, gives it a superficial )

        ( appearance of being right . . .       )

        (                                       )

        (                  -Thomas Paine 1776-  )

        ----------------------------------------

user9253

Joined: 28 Mar 2008
Posts: 1921
Location: Riley TWP Michigan

Posted: Thu May 29, 2008 7:12 am Post subject: Bridge Diodes use

 	  Quote: 	
		  Jeff,I assume that you have two power sources so that if one fails, then the other will carry the load.  If that is true, then each circuit should be designed to carry the full load. 	


 Joe

   [quote][b]

_________________
Joe Gores

nuckolls.bob(at)cox.net
Guest

Posted: Wed Jun 04, 2008 5:49 am Post subject: Bridge Diodes use

At 11:10 AM 5/29/2008 -0400, you wrote:



 	  Quote: 	
		  Jeff,



I assume that you have two power sources so that if one fails, then the 

other will carry the load.  If that is true, then each circuit should be 

designed to carry the full load.


	


   Correct.



   Bob . . .

jskiba(at)icosa.net
Guest

Posted: Wed Jun 04, 2008 6:42 am Post subject: Bridge Diodes use

Yes I do have two power sources (one with 90 amp capacity and one with 20 amp) and would plan on each being able to carry the load.However if Both are powered, from all the posts I read,the power source with the Higher  voltage will take ALL the load correct ?I ask since the two sources are not of the same capacity, aka during "normal ops"I would want to use the large capacity source to power the device connected to the diode.then if it fails the small device will pick up the load (with the help of extra battery capacity) until I can bring the load down on the smaller buss.Hope that makes sense :-)or am I missing something ?Jeff> --> AeroElectric-List message posted by: "Robert L. Nuckolls, III" > <nuckolls.bob(at)cox.net> > > At 11:10 AM 5/29/2008 -0400, you wrote: > >>Jeff, >> >>I assume that you have two power sources so that if one fails, then the >>other will carry the load. If that is true, then each circuit should be >>designed to carry the full load. > > Correct. > > Bob . . . >    [quote][b]

nuckolls.bob(at)cox.net
Guest

Posted: Wed Jun 04, 2008 6:25 pm Post subject: Bridge Diodes use

At 09:24 AM 6/4/2008 -0500, you wrote:

 	  Quote: 	
		  Yes I do have two power sources (one with 90 amp capacity and one with 20 

amp) and would plan on each being able to carry the load.



However if Both are powered, from all the posts I read,

the power source with the Higher  voltage will take ALL the load correct ?


	


   Correct.

 	  Quote: 	
		  I ask since the two sources are not of the same capacity, aka during 

"normal ops"

I would want to use the large capacity source to power the device 

connected to the diode.

then if it fails the small device will pick up the load (with the help of 

extra battery capacity) until I can bring the load down on the smaller buss.


	


 	  Quote: 	
		  Hope that makes sense 

or am I missing something ?


	


   I guess I don't understand your architecture. Did you

   start with one of the Z-figures?



   Bob . . .

Eric M. Jones

Joined: 10 Jan 2006
Posts: 565
Location: Massachusetts

Posted: Thu Jun 05, 2008 6:21 am Post subject: Re: Bridge Diodes use

 	  Quote: 	
		  Note 24. When you have critical loads that you would like

to accommodate with dual power sources, the 4-diode

bridge rectifier offers an easy to acquire, easy to mount,

easy to wire solution. The figure for this note illustrates

which terminals are used. Figure Z-19 shows one example

of how the device is used.

If your critical system draws more than 4 but less than 8

amps, the diode bridge should be mounted on a metallic

surface for heat sinking. If the loads are heavier, say 8 amps

up to the 25 or 30 amp rating of the device, perhaps a finned

heat sink is called for. (Aeroelectric Connection Note 24). 	




But let's be careful out there. Remember that a 25A Full Wave Bride is made up of four 12.5 Amp diodes. One must never make the mistake of thinking you can use it for 25A currents in a power-source-selecting circuit like Z-19. Bigger diodes are required.



"In times of rapid change, experience could be your worst enemy."

---Jean Paul Getty

_________________
Eric M. Jones
www.PerihelionDesign.com
113 Brentwood Drive
Southbridge, MA 01550
(508) 764-2072
emjones(at)charter.net

nuckolls.bob(at)cox.net
Guest

Posted: Thu Jun 05, 2008 10:15 am Post subject: Bridge Diodes use

At 07:21 AM 6/5/2008 -0700, you wrote:

 	  Quote: 	

 > Note 24. When you have critical loads that you would like

 > to accommodate with dual power sources, the 4-diode

 > bridge rectifier offers an easy to acquire, easy to mount,

 > easy to wire solution. The figure for this note illustrates

 > which terminals are used. Figure Z-19 shows one example

 > of how the device is used.

 > If your critical system draws more than 4 but less than 8

 > amps, the diode bridge should be mounted on a metallic

 > surface for heat sinking. If the loads are heavier, say 8 amps

 > up to the 25 or 30 amp rating of the device, perhaps a finned

 > heat sink is called for. Consult the membership of the

 > AeroElectric List for guidance in these special cases.

But let's be careful out there. Remember that a 25A Full Wave Bride is 

made up of four 12.5 Amp diodes. One must never make the mistake of 

thinking you can use it for 25A currents in a power-source-selecting 

circuit like Z-19. Bigger diodes are required.

   True! When I picked the bridge rectifier assembly for

   it's mechanical considerations (easy mounting, fast-on

   tab connections) the smallest device that came in that

   package was a 25A rated, full wave rectifier. 35A

   devices also come in that package. See:

http://aeroelectric.com/Pictures/Misc/s401-25.jpg

   It didn't occur to me then that a builder would find

   a practical application that would 'overload' the device

   electrically. As you've accurately pointed out, there

   are applications other than e-bus normal feed-path

   circuits that could begin to load one of these critters

   to it's rated limits. Referring to an exemplar

   data sheet . . .

http://aeroelectric.com/Mfgr_Data/Semiconductors/gbpc12.pdf

   . . . it's not clear but should be understood by

   the astute reader that only 1/2 of the components of

   a bridge rectifier are 'working' at any given time

   on alternate half-cycles of the incoming AC

   waveform. The data sheets don't offer a continuous

   duty rating for single devices. In practice, I suspect

   the individual devices can be used at more than 1/2

   the ratings for the total package. It's a matter of

   getting the heat out.

   From the neophyte builder's perspective, limiting

   continuous current on a single device to 1/2 the

   rated current for the whole device is a no-brainer

   conservative approach that will not disappoint.

   But keep in mind that ANY semi-conductor loaded

   to it's maximum continuous ratings dissipates

   heat . . . energy that must be removed through

   the mounting surfaces for the device. Further,

   no matter how large the ratings for any

   device - if the installation does not also take

   care of the heat generated. 10A of current flowing

   in a 100A rate part can smoke the device!

   There's nothing electrically 'magic' about the

   diode bridge rectifier illustrated above. It was

   originally suggested for its convenience of

   installation into systems that originally placed

   very low demands on its electrical abilities . . .

   no large concerns for rejecting heat. However,

   as one considers pushing the practical limits

   for this or any other device, close attention

   to ratings, limits and installation is called

   for.

   Thanks for bringing this up Eric.

   Bob . . .

longg(at)pjm.com
Guest

Posted: Fri Jun 06, 2008 6:03 pm Post subject: Bridge Diodes use

Wow, thanks - of course that is nice to know information before you buy the B & C 25A

 Diode and heat-sink which conservatively carries 12.5 A continuous.

 

 I know, read b4 you buy.

 

 Ok, then I need to go bigger. Can I melt the diode off the B & C heatsink to re-use?

 

 Looks like BR's are relatively cheap.

 

 All about Bridge Rectifiers

 

 http://hyperphysics.phy-astr.gsu.edu/Hbase/electronic/rectbr.html#c2

 

 

 Where to buy...

 

 $5.00 in the UK

 

 35A job

 

 http://www.reuk.co.uk/Buy-35A-Bridge-Rectifiers.htm

 

 Here's an 80A job for $6.50

 

 http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=360059397300

 

 With this 80A (conservatively 40A capacity unit) does my little 20A Ebus load need a heat-sink? Do I need  to fear voltage

 collapse if this thing is only getting 14V ?

 

 Glenn

 

 

 --

Eric M. Jones

Joined: 10 Jan 2006
Posts: 565
Location: Massachusetts

Posted: Tue Jun 10, 2008 6:16 am Post subject: Re: Bridge Diodes use

More on Diodes and FWBs: 



On B & C Diode Installation sheet they state that "Heatsink must dissipate 0.6 X Amps = Watts. Example For 20 Amp Essential Bus, 0.6 X 20 = 12 Watt Heatsink".



But this is wrong and points out a common fallacy regarding diodes:



With two resistors in parallel, the resistance of the combination is 1/2.

With two MosFets in parallel, the Rds(on) of the combination is 1/2

When using two (or any number of DIODES in parallel, the combination of the Vf's is still only Vf. B & C didn't understand this, and thus the dissipated wattage is twice what is stated. And remember--this bridge rectifier will ONLY handle 12.5 Amps with an infinite heatsink and NO margin when used in a source-selecting configuration.



This brings me back to my tired old point that anyone using regular diodes need to re-evaluate their choice. I have sold MANY HUNDREDS of PowerSchottkys to builders who understand this. For Z-19 etc. builders, check out the IXYS DSS 2x61-0045A Dual Power Schottky Rectifiers. 30A on each leg. I sell these with heatsink and Y-jumper, etc., but you can buy your own.



Before being concerned by the higher price--consider the savings in the much smaller, lighter heat sink, the higher performance, and the FAR LOWER COST OF OWNERSHIP. 



Someday my friend Bob will abandon this FWB stuff and go with Schottky Diodes. But I guess you can't get them at Radio Shack.



"In times of rapid change, experience could be your worst enemy."

---Jean Paul Getty

Power_Deuce_Schottky_ Manual.pdf

Description:

Download

Filename:

Power_Deuce_Schottky_ Manual.pdf

Filesize:

49.35 KB

Downloaded:

1124 Time(s)

_________________
Eric M. Jones
www.PerihelionDesign.com
113 Brentwood Drive
Southbridge, MA 01550
(508) 764-2072
emjones(at)charter.net

etienne.phillips(at)gmail
Guest

Posted: Tue Jun 10, 2008 8:34 am Post subject: Bridge Diodes use

 	  Quote: 	
		  >On B & C Diode Installation sheet they state that "Heatsink must dissipate 0.6 X Amps = Watts. Example For 20 Amp Essential Bus, 0.6 X 20 = 12 Watt Heatsink".

 	




In my experience, I have never come across a 12W heatsink. Nor a 1W or 1000W heatsink.



A heatsink is chosen appropriate to the temperature rise above ambient the casing of the device can tolerate, or what the spec-sheet says. A single machined-bolt head can dissipate 12W without any additional cooling, however only when it reaches 200 degrees C (thumbsuck). If you can only tolerate a 1degC increase in temperature, the heatsink required to dissipate 12W will be the size of a large desk.

 

So, a specsheet specifies the heatsinking requirements in [degrees above ambient]/[Watt], or specifies the maximum allowable temperature. So consider an ambient of 23degC, a maximum of 100degC, and we have calculated that the device will need to dissipate 10W. That gives us a requirement for a 7.7deg/Watt heatsink, which can be ordered accordingly.

 

I understand the calculations relating to dissipation of heat, and IIRC a Schottky diode, with a forward bias voltage of 0.2V rather than the 0.6~0.7V of a regular diode, dissipates 1/3 the energy for a given current. However, there are always other considerations that have not been touched in this discussion (such as thermal robustness, vibration tolerance, ease of mounting), some of which are the same for both Schottky diodes and for normal ones, and others which will draw very clear lines in the sand precluding one or both from certain applications.

  [quote][b]

longg(at)pjm.com
Guest

Posted: Tue Jun 10, 2008 9:05 am Post subject: Bridge Diodes use

Eric,

Thanks for the information. On Saturday I hooked up the B & C diode with

heat sink attached for a 6 hour load test (Z-19). I turned on 15 amps

worth of stuff to see if it would start smoking or something of the

sort. Initially it got 30-35 C but later cooled down below 30 C for the

duration of the run. No adverse behavior to report. If my load gets any

higher I'll be over the Schottky fence.



The Schottky device certainly looks more robust and if it runs cooler, I

may be the next fan.



 



--

kuffel(at)cyberport.net
Guest

Posted: Tue Jun 10, 2008 4:09 pm Post subject: Bridge Diodes use

Eric,



Am loath to get between you and Bob when discussing electrons but 

don't understand what you said about bridge diode heat 

dissipation.  If one diode is running 20 amps with a forward 

voltage of 0.6 then it generates 0.6 x 20 = 12 watts of heat.

If two diodes in parallel are carrying the same 20 amps then the 

current is split between the two diodes, with each having more or 

less half the load with a total of 20 amps between them.  So each 

diode is generating about half of 12 watts with a total 

generation of still only 12 watts not 24.



What am I missing?



Tom, AL7AU

mprather(at)spro.net
Guest

Posted: Tue Jun 10, 2008 4:47 pm Post subject: Bridge Diodes use

I'm with you Tom..



The only thing I'd add is that a diode has a pretty sharp knee in it's I-V

curve - where it's "resistance" goes from looking high (small change in I

for a large change in V) to where it's "resistance" looks low (large

change in I for small change in V).



If two unmatched devices are used in parallel, process differences between

them can cause one of the devices to turn on (forward bias) before the

other, and carry noticeable more current in that operating region.



Since significantly more voltage than the knee voltage is available in

this application, we can be fairly confident that both devices will be

fully forward biased, and the power drop should be fairly equal.



It's also likely that both devices come from the same fab lot (and wafer)

and so their electrical properties are probably essentially matched.

Regards,



Matt



 	  Quote: 	
		   

 <kuffel(at)cyberport.net>



 Eric,



 Am loath to get between you and Bob when discussing electrons but

 don't understand what you said about bridge diode heat

 dissipation.  If one diode is running 20 amps with a forward

 voltage of 0.6 then it generates 0.6 x 20 = 12 watts of heat.

 If two diodes in parallel are carrying the same 20 amps then the

 current is split between the two diodes, with each having more or

 less half the load with a total of 20 amps between them.  So each

 diode is generating about half of 12 watts with a total

 generation of still only 12 watts not 24.



 What am I missing?



 Tom, AL7AU

Eric M. Jones

Joined: 10 Jan 2006
Posts: 565
Location: Massachusetts

Posted: Tue Jun 10, 2008 5:57 pm Post subject: Re: Bridge Diodes use

 	  Quote: 	
		  Eric, 



Am loath to get between you and Bob when discussing electrons but 

don't understand what you said about bridge diode heat 

dissipation. If one diode is running 20 amps with a forward 

voltage of 0.6 then it generates 0.6 x 20 = 12 watts of heat. 

If two diodes in parallel are carrying the same 20 amps then the 

current is split between the two diodes, with each having more or 

less half the load with a total of 20 amps between them. So each 

diode is generating about half of 12 watts with a total 

generation of still only 12 watts not 24. 

What am I missing? 

Tom, AL7AU 	




Tom et al.



If either one or two or a hundred conventional diodes is sharing the load, they still dissipate Vf X A. Where A is the current through the device; in this case the Full Wave Bridge (FWB) or Schottky Module.



The important point is that Vf is not reduced by putting parts in parallel. (But it is increased by putting parts in series....) 



(True, each diode carries less current, but the package is what we bolt to the heat sink. In fact you can distribute the heat dissipation by distributing the parts, and this is done on some designs, since small parts have greater surface area per volume, so they might not need a heat sink at all.)



 	  Quote: 	
		  If one diode is running 20 amps with a forward 

voltage of 0.6 then it generates 0.6 x 20 = 12 watts of heat. 	




If pigs had wings!....The error I pointed out is the claim that a B & C FWB, when wired so that two diodes are parallel, would have a Vf of 0.6. It does not; the Vf is often 1.2 maybe, and someone incorrectly assumed that the diodes being in parallel made the Vf sum 0.6. The very best conventional FWB diodes are 0.9Vf at 10A for each diode (you can't buy these at Radio Shack), so it dissipates (2X0.9X10A=) 18W VERY BEST CASE., and 24W worst case if the diodes are Vf=1.2.



By the way, FWB packages are just four diodes wired up to four terminals and epoxy potted in a little box. You could make your own, but it's easier to abandon the concept entirely.



Hope this doesn't scramble it more.



For a successful technology, reality must take precedence over public

relations, for Nature cannot be fooled.

      -- Richard P. Feynman



[/quote]

_________________
Eric M. Jones
www.PerihelionDesign.com
113 Brentwood Drive
Southbridge, MA 01550
(508) 764-2072
emjones(at)charter.net

sportav8r(at)gmail.com
Guest

Posted: Tue Jun 10, 2008 6:34 pm Post subject: Bridge Diodes use

So you're saying that any given diode (of the four) in a conventional, inexpensive  potted FWB rectifier has a Vf typ of 1.2V?  That's news - very bad news.  I always thought 0.6V was closer to average, per Si diode, and about half that for Ge diodes.  Learned something today.

-Bill B

On Tue, Jun 10, 2008 at 9:57 PM, Eric M. Jones <emjones(at)charter.net (emjones(at)charter.net)> wrote:

[quote] --> AeroElectric-List message posted by: "Eric M. Jones" <emjones(at)charter.net (emjones(at)charter.net)>

 > Eric,

 >

 > Am loath to get between you and Bob when discussing electrons but

 > don't understand what you said about bridge diode heat

 > dissipation. If one diode is running 20 amps with a forward

 > voltage of 0.6 then it generates 0.6 x 20 = 12 watts of heat.

 > If two diodes in parallel are carrying the same 20 amps then the

 > current is split between the two diodes, with each having more or

 > less half the load with a total of 20 amps between them. So each

 > diode is generating about half of 12 watts with a total

 > generation of still only 12 watts not 24.

 > What am I missing?

 > Tom, AL7AU

 Tom et al.

 If either one or two or a hundred conventional diodes is sharing the load, they still dissipate Vf X A. Where A is the current through the device; in this case the Full Wave Bridge (FWB) or Schottky Module.

 The important point is that Vf is not reduced by putting parts in parallel. (But it is increased by putting parts in series....)

 (True, each diode carries less current, but the package is what we bolt to the heat sink. In fact you can distribute the heat dissipation by distributing the parts, and this is done on some designs, since small parts have greater surface area per volume, so they might not need a heat sink at all.)

 > If one diode is running 20 amps with a forward

 > voltage of 0.6 then it generates 0.6 x 20 = 12 watts of heat.

 If pigs had wings!....The error I pointed out is the claim that a B & C FWB, when wired so that two diodes are parallel, would have a Vf of 0.6. It does not; the Vf is often 1.2 maybe, and someone incorrectly assumed that the diodes being in parallel made the Vf sum 0.6. The very best conventional FWB diodes are 0.9Vf at 10A for each diode (you can't buy these at Radio Shack), so it dissipates (2X0.9X10A=) 18W VERY BEST CASE., and 24W worst case if the diodes are Vf=1.2.

 By the way, FWB packages are just four diodes wired up to four terminals and epoxy potted in a little box. You could make your own, but it's easier to abandon the concept entirely.

 Hope this doesn't scramble it more.

 For a successful technology, reality must take precedence over public

 relations, for Nature cannot be fooled.

       -- Richard P. Feynman

 --------

 Eric M. Jones

 www.PerihelionDesign.com

 113 Brentwood Drive

 Southbridge, MA 01550

 (508) 764-2072

 emjones(at)charter.net (emjones(at)charter.net)

 Read this topic online here:

 http://forums.matronics.com/viewtopic.php?p=187299#187299

 [b]

mprather(at)spro.net
Guest

Posted: Tue Jun 10, 2008 7:03 pm Post subject: Bridge Diodes use

I believe a normal Si diode has a Vf around 0.6V - 0.7V.



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



A bridge rectifier would normally be wired so that two diodes are in

series.  However, many of these bridge rectifiers are built to allow

access to all four nodes of the circuit.



http://www.vishay.com/docs/88612/gbpc12.pdf



This datasheet indeed shows a Vf of 1.1V - used as a rectifier.  I'm

pretty sure this is measured across the + and - terminals - two diodes in

series.



I believe it has been proposed that these rectifiers be used to feed the

E-bus by shorting across two of the diodes and paralleling the other two,

and a Vf = 0.6V.



I missed the original reference to the B&C guidance that paralleling

dropped the Vf.  Please post that if you have it..

Regards,



Matt-



 	  Quote: 	
		   

 <emjones(at)charter.net>

> Eric,

>

> Am loath to get between you and Bob when discussing electrons but

> don't understand what you said about bridge diode heat

> dissipation. If one diode is running 20 amps with a forward

> voltage of 0.6 then it generates 0.6 x 20 = 12 watts of heat.

> If two diodes in parallel are carrying the same 20 amps then the

> current is split between the two diodes, with each having more or

> less half the load with a total of 20 amps between them. So each

> diode is generating about half of 12 watts with a total

> generation of still only 12 watts not 24.

> What am I missing?

> Tom, AL7AU

 Tom et al.



 If either one or two or a hundred conventional diodes is sharing the load,

 they still dissipate Vf X A. Where A is the current through the device; in

 this case the Full Wave Bridge (FWB) or Schottky Module.



 The important point is that Vf is not reduced by putting parts in

 parallel. (But it is increased by putting parts in series....)



 (True, each diode carries less current, but the package is what we bolt to

 the heat sink. In fact you can distribute the heat dissipation by

 distributing the parts, and this is done on some designs, since small

 parts have greater surface area per volume, so they might not need a heat

 sink at all.)

> If one diode is running 20 amps with a forward

> voltage of 0.6 then it generates 0.6 x 20 = 12 watts of heat.

 If pigs had wings!....The error I pointed out is the claim that a B & C

 FWB, when wired so that two diodes are parallel, would have a Vf of 0.6.

 It does not; the Vf is often 1.2 maybe, and someone incorrectly assumed

 that the diodes being in parallel made the Vf sum 0.6. The very best

 conventional FWB diodes are 0.9Vf at 10A for each diode (you can't buy

 these at Radio Shack), so it dissipates (2X0.9X10A=) 18W VERY BEST CASE.,

 and 24W worst case if the diodes are Vf=1.2.



 By the way, FWB packages are just four diodes wired up to four terminals

 and epoxy potted in a little box. You could make your own, but it's easier

 to abandon the concept entirely.



 Hope this doesn't scramble it more.



 For a successful technology, reality must take precedence over public

 relations, for Nature cannot be fooled.

       -- Richard P. Feynman

 --------

 Eric M. Jones

 www.PerihelionDesign.com

 113 Brentwood Drive

 Southbridge, MA 01550

 (508) 764-2072

 emjones(at)charter.net





 Read this topic online here:



 http://forums.matronics.com/viewtopic.php?p=187299#187299

kuffel(at)cyberport.net
Guest

Posted: Wed Jun 11, 2008 12:29 am Post subject: Bridge Diodes use

Eric,



Still don't follow you.



I said:

<< If one diode is running 20 amps with a forward voltage of 0.6 

then it generates 0.6 x 20 = 12 watts of heat. If two diodes in 

parallel are carrying the same 20 amps then the current is split 

between the two diodes .. each diode is generating about half of 

12 watts with a total generation of still only 12 watts not 24. >>



You said:

<< If either one or two or a hundred conventional diodes is 

sharing the load, they still dissipate Vf X A. >>



Agreed.  But each diode is carrying a smaller fraction of the 

same total amps so the total example heat generated remains 12 

watts.  Put algebraically:



Let Vf x A = 12 Watts

Let 3 diodes in parallel carry the same total current so

Vf1 = Vf2 = Vf3 = Vf

a1 + a2 + a3 = A

Then

(Vf1 x a1) + (Vf2 x a2) + (Vf3 x a3) = Vf x (a1 + a2 + a3)

  = Vf x A = 12 watts



<< Vf is not reduced by putting parts in parallel. >>



Never said it did reduce Vf.  (I've always used 0.7V for silicon 

and 0.3V for germanium in my minimal design work but a Vf of 0.6V 

is close enough also for this discussion.)  I did say it reduced 

the current carried by each individual part.



<< but the package is what we bolt to the heat sink. >>



But the package in your example is still only generating 12 watts 

of heat, not 24.  See below.



<< claim that a B & C FWB, when wired so that two diodes are 

parallel, would have a Vf of 0.6. It does not; the Vf is often 

1.2 maybe, and someone incorrectly assumed that the diodes being 

in parallel made the Vf sum 0.6. The very best conventional FWB 

diodes are 0.9Vf at 10A for each diode (you can't buy these at 

Radio Shack), so it dissipates (2X0.9X10A=) 18W VERY BEST CASE., 

and 24W worst case if the diodes are Vf=1.2. >>



As Matt says, a Vf of 1.2 is only true when the FWB is wired up 

as a full wave rectifier.  Then there are indeed two diodes 

working in *series* for each half cycle with a Vf of about 1.3V.



Perhaps, you might be referring to the fact the forward voltage 

drop across a single diode increases from the turn-on voltage 

value with increasing current.  This can get above 1 volt at high 

currents.  But your use of 0.6V clearly made the discussion at 

the turn-on voltage point not the Vf at high current.  The 

situation is the same whether we talk about Vf at turn-on current 

or Vf at high current as long as we remain consistent.



And as an aside, putting diodes in parallel brings us lower on 

the current scale for each device and closer to its turn-on 

voltage.  In other words, putting diodes in parallel lowers 

(slightly) the total heat generated for the same total current.



In parallel, each diode still has it's original Vf and since they 

are in parallel they don't add voltages.  Whether there is 1 or 

100 diodes in the package, they are still only carrying the 

original example 20 amps total with a voltage drop between the 

input terminal and the output terminal of one Vf since there is 

only one diode drop between the terminals no matter how many 

diodes are wired in parallel in between.



<< FWB ... You could make your own, but it's easier to abandon 

the concept entirely. >>



Not claiming to be a FWB fan for this application.  Just don't 

see how running 20 amps total across one diode drop doubles the 

heat generated when another diode is added to share the same 

current.  Under this logic, 8 diodes in parallel would generate 

96 watts, 16 would generate 192 watts, etc. which is obviously 

not what you mean to imply.



Basically, do not see how wiring diodes in parallel causes their 

forward voltage drops to add together as you seem to be saying.



Tom, AL7AU

bob(at)bob-white.com
Guest

Posted: Wed Jun 11, 2008 6:07 am Post subject: Bridge Diodes use

If you look up single silicon diodes in the 10 A to 20 A current rating

on Mouser or Digikey, you will find the Vf is 1.1 Volts or higher for

most of them. For example, a 1N3210 (a 15 A unit) is 1.5 V at 15 amps.

Look up a 1N4007 rated at 1 amp.  Vf is 1.1V.  Eric is right on the

money on this. He is just saying you can't use 0.6 Volts for Vf because

you have two 1.2 volt devices in parallel.



Bob W.

On Wed, 11 Jun 2008 02:25:10 -0600

The Kuffels <kuffel(at)cyberport.net> wrote:



 	  Quote: 	
		   

 

 Eric,

 

 Still don't follow you.

 

 I said:

 << If one diode is running 20 amps with a forward voltage of 0.6 

 then it generates 0.6 x 20 = 12 watts of heat. If two diodes in 

 parallel are carrying the same 20 amps then the current is split 

 between the two diodes .. each diode is generating about half of 

 12 watts with a total generation of still only 12 watts not 24. >>

 

 You said:

 << If either one or two or a hundred conventional diodes is 

 sharing the load, they still dissipate Vf X A. >>

 

 Agreed.  But each diode is carrying a smaller fraction of the 

 same total amps so the total example heat generated remains 12 

 watts.  Put algebraically:

 

 Let Vf x A = 12 Watts

 Let 3 diodes in parallel carry the same total current so

 Vf1 = Vf2 = Vf3 = Vf

 a1 + a2 + a3 = A

 Then

 (Vf1 x a1) + (Vf2 x a2) + (Vf3 x a3) = Vf x (a1 + a2 + a3)

   = Vf x A = 12 watts

 

 << Vf is not reduced by putting parts in parallel. >>

 

 Never said it did reduce Vf.  (I've always used 0.7V for silicon 

 and 0.3V for germanium in my minimal design work but a Vf of 0.6V 

 is close enough also for this discussion.)  I did say it reduced 

 the current carried by each individual part.

 

 << but the package is what we bolt to the heat sink. >>

 

 But the package in your example is still only generating 12 watts 

 of heat, not 24.  See below.

 

 << claim that a B & C FWB, when wired so that two diodes are 

 parallel, would have a Vf of 0.6. It does not; the Vf is often 

 1.2 maybe, and someone incorrectly assumed that the diodes being 

 in parallel made the Vf sum 0.6. The very best conventional FWB 

 diodes are 0.9Vf at 10A for each diode (you can't buy these at 

 Radio Shack), so it dissipates (2X0.9X10A=) 18W VERY BEST CASE., 

 and 24W worst case if the diodes are Vf=1.2. >>

 

 As Matt says, a Vf of 1.2 is only true when the FWB is wired up 

 as a full wave rectifier.  Then there are indeed two diodes 

 working in *series* for each half cycle with a Vf of about 1.3V.

 

 Perhaps, you might be referring to the fact the forward voltage 

 drop across a single diode increases from the turn-on voltage 

 value with increasing current.  This can get above 1 volt at high 

 currents.  But your use of 0.6V clearly made the discussion at 

 the turn-on voltage point not the Vf at high current.  The 

 situation is the same whether we talk about Vf at turn-on current 

 or Vf at high current as long as we remain consistent.

 

 And as an aside, putting diodes in parallel brings us lower on 

 the current scale for each device and closer to its turn-on 

 voltage.  In other words, putting diodes in parallel lowers 

 (slightly) the total heat generated for the same total current.

 

 In parallel, each diode still has it's original Vf and since they 

 are in parallel they don't add voltages.  Whether there is 1 or 

 100 diodes in the package, they are still only carrying the 

 original example 20 amps total with a voltage drop between the 

 input terminal and the output terminal of one Vf since there is 

 only one diode drop between the terminals no matter how many 

 diodes are wired in parallel in between.

 

 << FWB ... You could make your own, but it's easier to abandon 

 the concept entirely. >>

 

 Not claiming to be a FWB fan for this application.  Just don't 

 see how running 20 amps total across one diode drop doubles the 

 heat generated when another diode is added to share the same 

 current.  Under this logic, 8 diodes in parallel would generate 

 96 watts, 16 would generate 192 watts, etc. which is obviously 

 not what you mean to imply.

 

 Basically, do not see how wiring diodes in parallel causes their 

 forward voltage drops to add together as you seem to be saying.

 

 Tom, AL7AU

 


	




-- 

N93BD - Rotary Powered BD-4 - http://www.bob-white.com

3.8 Hours Total Time and holding

Cables for your rotary installation - http://roblinstores.com/cables/

Eric M. Jones

Joined: 10 Jan 2006
Posts: 565
Location: Massachusetts

Posted: Wed Jun 11, 2008 7:09 am Post subject: Re: Bridge Diodes use

 	  Quote: 	
		  If you look up single silicon diodes in the 10 A to 20 A current rating 

on Mouser or Digikey, you will find the Vf is 1.1 Volts or higher for 

most of them. For example, a 1N3210 (a 15 A unit) is 1.5 V at 15 amps. 

Look up a 1N4007 rated at 1 amp. Vf is 1.1V. Eric is right on the 

money on this. He is just saying you can't use 0.6 Volts for Vf because 

you have two 1.2 volt devices in parallel. 

Bob W. 	




Yeah...What HE said...! Thanks, Bob White.



I can't in any reasonable time answer some previous posts, but there are a couple issues that seem to confuse some people:



The FWB when used as Bob N. and B & C recommends uses two diodes in parallel. The watts dissipated by the part at 20A is WORST CASE 1.2 x 20=24Watts.



The Vf of a diode is a strongly influenced by current. A diode can be 0.7 Vf at zero Amps, and 1.0 Vf at 10A and 1.2 Vf at 12.5. For this you have to read the data sheet. What?! Radio Shack doesn't supply one?! Then just guess and hope you're right. Double up on the life insurance, too.



 

 "In times of rapid change, experience could be your worst enemy."

     --Jean Paul Getty

_________________
Eric M. Jones
www.PerihelionDesign.com
113 Brentwood Drive
Southbridge, MA 01550
(508) 764-2072
emjones(at)charter.net

Display posts from previous:

	Matronics Email Lists Forum Index -> AeroElectric-List	All times are GMT - 8 Hours Goto page 1, 2 Next
Page 1 of 2

Jump to:

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum