high power zener diodes

Thread Starter

setsquar

Joined Oct 11, 2012
34
been looking for some high power zener diodes and found the BZY91C13
which is 100 watt but the data sheet says 2amp ?
Can anyone explain that please
thanks for looking
 

OBW0549

Joined Mar 2, 2015
3,566
been looking for some high power zener diodes and found the BZY91C13
which is 100 watt but the data sheet says 2amp ?
Can anyone explain that please
That 2.0 Amp is not the maximum allowable current; it is the Zener test current-- i.e., the current at which the Zener voltage is tested and specified.
 

ian field

Joined Oct 27, 2012
6,539
been looking for some high power zener diodes and found the BZY91C13
which is 100 watt but the data sheet says 2amp ?
Can anyone explain that please
thanks for looking
The old British motorcycles were regulated by a dirty great Zener, which I think was 100W.

With minimal load and fully charged battery, they may have been called on to sink as much as 13A.
 

dendad

Joined Feb 20, 2016
3,625
You could also use a zener and power transistor to get a similar result, and it will probably be a lot cheaper. How accurate do you need it?
 

crutschow

Joined Mar 14, 2008
25,701
Another option to emulate an accurate, high-power zener is to use an inexpensive TL431 programmable shunt regulator with a power P-MOSFET current booster as shown in the LTspice simulation below.
The voltage divider R1 and R2 can be adjusted to give any "zener" voltage from 2.5V to 36V (show set for 13V in the simulation with a simulated current of 0-10A).
Their value is selected to give 2.5V at the TL431 control input for the desired clamp voltage.

A pot can also be used to adjust the voltage.

This voltage is more stable and accurate than a typical zener, with a much lower dynamic impedance and temperature coefficient.

Of course, the transistor must be mounted on a heatsink to handle its dissipation, which equals the zener voltage times the current.

upload_2018-5-13_22-22-36.png
 
Last edited:

ian field

Joined Oct 27, 2012
6,539
Another option to emulate an accurate, high-power zener is to use an inexpensive TL431 programmable shunt regulator with a power P-MOSFET current booster as shown in the LTspice simulation below.
The voltage divider R1 and R2 can be adjusted to give any "zener" voltage from 2.5V to 36V (show set for 13V in the simulation with a simulated current of 0-10A).
Their value is selected to give 2.5V at the TL431 control input for the desired clamp voltage.

A pot can also be used to adjust the voltage.

This voltage is more stable and accurate than a typical zener, with a much lower dynamic impedance and temperature coefficient.

Of course, the transistor must be mounted on a heatsink to handle its dissipation, which equals the zener voltage times the current.

View attachment 152464
You need enough voltage headroom to accommodate VGSthr - the TL431 can carry 100mA as long as you don't exceed dissipation, so there's no particular advantage in using a MOSFET over using a BJT.
 

crutschow

Joined Mar 14, 2008
25,701
the TL431 can carry 100mA as long as you don't exceed dissipation, so there's no particular advantage in using a MOSFET over using a BJT.
If you want the BJT to carry more than a couple amps, it would need to be either a Darlington stage or a MOSFET.

For low voltage equivalent zener voltage (6V or so) you would need a logic-level MOSFET with a Vth(max) of ≤2V.
 

ian field

Joined Oct 27, 2012
6,539
If you want the BJT to carry more than a couple amps, it would need to be either a Darlington stage or a MOSFET.

For low voltage equivalent zener voltage (6V or so) you would need a logic-level MOSFET with a Vth(max) of ≤2V.
Everythings a load of compromises - a Darlington can need as much voltage as a LL MOSFET. For 100w equivalency, a power MOSFET isn't particularly outstanding.
 

crutschow

Joined Mar 14, 2008
25,701
For the lowest equivalent Zener voltage setting, you could use a Sziklai (complementary) Darlington Pair in place of a MOSFET, which only requires about 0.7V of headroom added to the ≈2V minimum required by the TL431.
That should allow setting to the minimum of 2.5V for the TL431.
The 2N3055 has a 10A maximum rating with the circuit operating limit being about 5A.
Edit: At higher current and voltage levels the dissipation of the 2N2907 will become excessive so a higher power transistor which can also be mounted on a heatsink, such as a TIP32, should be used for Q2.

upload_2018-5-14_14-5-56.png
 
Last edited:

ian field

Joined Oct 27, 2012
6,539
For the lowest equivalent Zener voltage setting, you could use a Sziklai (complementary) Darlington Pair in place of a MOSFET, which only requires about 0.7V of headroom added to the ≈2V minimum required by the TL431.
The 2N3055 has a 10A maximum rating.

View attachment 152491
You can make a Sziklai hybrid with the TL431 - but probably not for that much current. I use it in the shunt regulators for my DIY E-cig chargers.
 

Thread Starter

setsquar

Joined Oct 11, 2012
34
Background,
in the late 1960/70s British bikes had a permanent magnet alternator
when they went from 6volt to 12v lucas came up with a simple Zener diode to regulate the voltage between about 13 and 14.7 volts, its simple and worked ok
these are no longer available, so I would like to make the simplest circuit possible to emulate the original
alternator excess output with engine running and lights on is about 20 watts (I always ride with lights on)
without lights its nearer 60+ watts
I found this circuit on the web, is it viable if so what value and spec of components would I need
thanks for your help
http://imgbox.com/h6ntjV5S"

ps sorry about the photos took a while to figer out how to do it

MOD:
h6ntjV5S_o.jpg.png

 
Last edited by a moderator:

crutschow

Joined Mar 14, 2008
25,701
D1 zener voltage would be about 0.7V less than the voltage you want to regulate (probably about 14V for a 12V battery, so a zener voltage of 13.3V.
You would use a standard value of 13V±5%.

The value of R1 is not critical, so 10kΩ should work fine.

For a more accurate voltage setting and regulation, you could use one of the TL431 circuits I posted above.
 

RichardO

Joined May 4, 2013
2,271
Background,
in the late 1960/70s British bikes had a permanent magnet alternator...
... so I would like to make the simplest circuit possible to emulate the original
alternator excess output with engine running and lights on is about 20 watts (I always ride with lights on)
without lights its nearer 60+ watts
I found this circuit on the web, is it viable if so what value and spec of components would I need
thanks for your help
D1 zener voltage would be about 0.7V less than the voltage you want to regulate (probably about 14V for a 12V battery, so a zener voltage of 13.3V.
You would use a standard value of 13V±5%.

The value of R1 is not critical, so 10kΩ should work fine.

For a more accurate voltage setting and regulation, you could use one of the TL431 circuits I posted above.
You can probably use the 2N3055 (or the TIP3055) for the NPN in your circuit. Either one will have to be on a large heat sink. I assume that you already have a heat sink that the power Zener was mounted on. You might be able to make it work for the '3055.

Since getting a 13 volt Zener might be hard to find, you might want to put 1 or 2 1N4001 rectifier diodes in series with a 1N4742A Zener diode. (The diodes will be in the opposite polarity to that of the Zener).
 

crutschow

Joined Mar 14, 2008
25,701
Here's an LTspice simulation of the zener-transistor circuit.

I used common 5.6V and 7.5V zeners in series to get the desired output voltage.

I added a second transistor to form a Darlington to reduce the zener power, which was otherwise over 2W each.
Note that Q1 is dissipating 66W and Q2 is dissipating 4.7W @ 5A, so both need to be on a heatsink.

Q2 could also be a smaller transistor, such as a TIP31.

The simulated regulated voltage is 13.9V @ 2A and 14.3V @ 5A.
Thes can vary at least ±5% due to component tolerances, which is a problem with such a simple circuit.
You may have to select different zener voltages to get the desired output in the real circuit.
The circuits using a TL431 would have a much tighter tolerance and regulation.

upload_2018-5-15_10-35-59.png
 
Last edited:

Thread Starter

setsquar

Joined Oct 11, 2012
34
Thanks i will give the 3055 a try
there is NO battery fitted
very low volts at idle speed with lights on
so i dont think any TL431 IC type circuits will not work

but thanks everyone for the help
 
Top