Author Topic: Zener diode dissipation for protection.  (Read 2298 times)

0 Members and 1 Guest are viewing this topic.

Offline davelectronicTopic starter

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: gb
  • Life's too short.
Zener diode dissipation for protection.
« on: May 13, 2021, 11:19:11 pm »
I'm trying to work out if I can increase a zener diodes power dissipation ability for circuit over voltage protection. I don't have room in two power supply projects to add in a crowbar circuit. So I wanted to use a fuse and zener diode, just incase the series pass transistors where to fail short circuit. But the zener diodes I have are rated at 16 Volts and 5 watts power rating. I think if the 16 Volt 5 watt zener does reverse conduct it would probably burn up before it blows the fuse. I might not be right, but I think I need a series current limiting resistor for the zener diode. Really I don't want to power up radio equipment unless I have some over voltage protection. Years ago I had a retail Altai PSU push out over 20 Volts and destroy a radio. So I need a solution that doesn't take up much space inside the power supply case. Does that sound right about using a resistor to prevent the zener diode burning up ? The maximum current that the power supply will deliver is 12 Amps, at 12.60 Volts DC. I need a solution if one or more series pass transistors fail short circuit. Any ideas and help appreciated. Thanks for reading.
 

Offline davelectronicTopic starter

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: gb
  • Life's too short.
Re: Zener diode dissipation for protection.
« Reply #1 on: May 14, 2021, 01:40:03 am »
Just found something, but don't know if its useful or not, that was using a BJT to shunt the zener diode. But its power rating was raised by the beta of the transistor. Any thoughts if that's possible for a 12 Amp 12 volt power supply over voltage protection ?
 

Online David Hess

  • Super Contributor
  • ***
  • Posts: 16510
  • Country: us
  • DavidH
Re: Zener diode dissipation for protection.
« Reply #2 on: May 14, 2021, 01:41:23 am »
I will assume that you mean a TVS diode because they are designed to have overload capability.  True zener/avalanch regulator diodes are not as robust.

A big TVS diode can work to blow a fuse but needs to be sized properly, or the TVS diode will short, although that of course will blow the fuse also.

I have done it with a zener or TVS diode and a big bipolar power transistor when I needed a higher power rating than a zener or TVS diode will allow.  But if you have room for a transistor and zener, then you have room for an SCR and zener.

« Last Edit: May 14, 2021, 01:43:36 am by David Hess »
 
The following users thanked this post: davelectronic

Offline davelectronicTopic starter

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: gb
  • Life's too short.
Re: Zener diode dissipation for protection.
« Reply #3 on: May 14, 2021, 03:20:42 pm »
It was a zener diode and fuse I was referring to, I hadn't given TVS diode any thoughts. I will snap a picture inside the power supply to give an idea on amount of space I have inside the PSU. I've had a few issues with a crowbar circuit firing randomly, it's a real pain it kept doing this. I had included the capacitor to stop it firing randomly if the voltage spiked. I'm just looking for an over voltage protection circuit that just works, and figured a zener diode and fuse could fill that brief, but quickly realised the zener diode power rating would be a problem. So was looking for a solution to increase the 5 watt zener diodes current handling rating.
 

Online David Hess

  • Super Contributor
  • ***
  • Posts: 16510
  • Country: us
  • DavidH
Re: Zener diode dissipation for protection.
« Reply #4 on: May 14, 2021, 04:39:17 pm »
It was a zener diode and fuse I was referring to, I hadn't given TVS diode any thoughts.

TVS diodes are zener diodes that are processed and specified to handle surge currents.  They no longer make big zener diodes like they used to because they were replaced with TVS diodes.  There is no reason not to use a TVS diode in place of a zener diode.

The processing difference is the same as that used to make avalanche rated rectifiers.  The junction is made more uniform so that hot spots do not form which would allow localized thermal runaway.
 
The following users thanked this post: davelectronic

Offline Tomorokoshi

  • Super Contributor
  • ***
  • Posts: 1212
  • Country: us
Re: Zener diode dissipation for protection.
« Reply #5 on: May 14, 2021, 05:41:03 pm »
1. How much power do you need to supply to the radio project? The full 12A at 12.6V, or something less?

2. Look at something like the 8W P600 package TVS diodes.

3. If you use a TVS, the datasheets won't give precise guidance. You may need to test. 5KP11A, 5KP12A, or 5KP13A might be close.

4. You can put these in parallel. Buy extra and match them. If they are close because of the temperature and resistance coefficient they will load share reasonably.

5. Heat sink if necessary.
 
The following users thanked this post: davelectronic

Offline davelectronicTopic starter

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: gb
  • Life's too short.
Re: Zener diode dissipation for protection.
« Reply #6 on: May 14, 2021, 11:09:21 pm »
Yes 12.60 Volts at 12 Amps would be the highest power drain scenario. More typical would be around 8 Amps at 12.60 volts. The voltage at 12 Amps drops a little to 12.35 Volts but it steady. I've never used TVS diodes, I know they clamp transients I think, but unsure how they would see an over voltage event of the full 20 Volt input to the regulator / series pass transistors circuit. As I understand it, if a transistors fail short circuit, that full 20 Volts would make its way to the radio transceiver and linear amplifier set up.
 

Offline davelectronicTopic starter

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: gb
  • Life's too short.
Re: Zener diode dissipation for protection.
« Reply #7 on: May 15, 2021, 09:28:37 pm »
Not wanting to start another thread, is there someone that can take a glance over this over voltage protection circuit please. I might have to find an alternative SCR if the one in the circuit is unavailable. Any help appreciated.
 

Offline Jwillis

  • Super Contributor
  • ***
  • Posts: 1673
  • Country: ca
Re: Zener diode dissipation for protection.
« Reply #8 on: May 16, 2021, 09:35:38 am »
Not wanting to start another thread, is there someone that can take a glance over this over voltage protection circuit please. I might have to find an alternative SCR if the one in the circuit is unavailable. Any help appreciated.


First off that TVS is only rated with a Steady State Power Dissipation on Infinite Heat Sink  at TL=75ÂșC of only 6.5W At 13.8 V that's  only 470mA current . It will burn up long before the fuses do . TVS diode high current capabilities are rated for current spikes of only a few micro seconds not for long duration's . Look up SCR crowbar circuit and you'll find that there are no TVS diodes .
 Only a zener , resistor and SCR in a basic voltage clamp . The concept is when the voltage gets high enough to reach the breakdown voltage of the zener then it begins to conduct hence raising the voltage of the gate . This switches on the SCR there by clamping the over voltage and current to ground . The resistor limits the current through the zener .
 
The following users thanked this post: davelectronic

Offline davelectronicTopic starter

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: gb
  • Life's too short.
Re: Zener diode dissipation for protection.
« Reply #9 on: May 16, 2021, 01:36:05 pm »
Thanks for replying, I see the TVS diode is not for prolonged voltage spiked. I'm finding some information on the web conflicting, I've found what I thought would be an ideal circuit. But there is two versions of this crowbar circuit, but one has less capacitors and a single current limit resistor, and the other drawn in the same configuration with more capacitors and two resistors. One I assume is the SCR gate current limit resistor, and the other I think is a resistor to limit current through the zener diode. I don't know which is more suitable for protection of over voltage imposed to radio equipment running on 12 volts to 13.80 Volts power source. I've added the two crowbar circuits, but the only way I would know which one is suitable is to construct both and test both. Also I found another crowbar circuit, but that uses a TVS diode again.
 

Offline Jwillis

  • Super Contributor
  • ***
  • Posts: 1673
  • Country: ca
Re: Zener diode dissipation for protection.
« Reply #10 on: May 16, 2021, 07:14:49 pm »
Zener diodes have a maximum current and a minimum breakdown current . The breakdown current is the minimum current to "turn on " the zener .

Minimum resistor value: Rmin=(Vs-Vz)/Izmax
Maximum resistor value: Rmax=(Vs-Vz)/Izmin

Where Vs is the maximum source voltage . Vz is the zener voltage . Izmax is the zener maximum current , Izmin is the zener minimum current

Usually the rated currents for a zener will be in the datasheet.

You would choose a resistor within those parameters based on current requirements of the SCR gate . The SCR gate has a maximum gate current and a minimum gate current to trigger SCR . Something you need to play around with since zeners have a fairly wide range of error . Plus or minus 5 percent in most cases.

Zeners also drift with temperature . Under 5.1V they drift down in voltage as temperture increases (negative coefficient) and drift up in voltage above 5.1V as temperature increases (positive coefficient)
You can also put several zeners in series to increase power rating and reduce drift . For example if you put Two 1W 6V zeners in series you will have a 2W 12 V zener equivalent with less temperature drift than one single  12V zener .

The caps are  to ensure the SCR does not trigger with short voltage spikes. Like when you turn on the power supply .

The SCR Voltage clamp is a crude way to protect your circuits but it does work . Don't expect it to have very tight parameters .




 
The following users thanked this post: davelectronic

Offline davelectronicTopic starter

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: gb
  • Life's too short.
Re: Zener diode dissipation for protection.
« Reply #11 on: May 16, 2021, 07:48:38 pm »
Thank you for explaining these parameters, I knew some of them. Sometimes a data sheet does get a bit overwhelming, maths isn't my strongest subject, I do struggle a bit with it. This afternoon I've had a play with a thyristor zener diode and 470 ohm half watt resistor, oh and a 10 Amp 20mm fuse. The load was a single 50 watt halogen lamp. It works, but there where a few bumps along the way. First I tried a TYN1225 SCR and a zener diode 1N5351B and a 470 ohm resistor also 0.1 uf capacitor. The SCR just died if I raised the voltage at the point the zener reverse voltage started conducting. The zener survived, the fuse didn't blow. I changed the SCR to a 2N6509G, this combination worked, I tested it over and over blowing 10 Amp fuses just to see if the components where survivable. They where, one thing I noted is the 50 watt lamp load had the briefest of glows, but didn't distroy the lamp filermant. Almost like the very shortest of delays, but after about 6 blown fuses the lamp survived each over voltage scenario. I powered it back up through the crowbar circuit with 12 Volts, and it was fine. I don't know the issue with the TYN1225 that died twice, but the 2N6509G SCR and the zener diode 1N5351B worked ok I think. I'm unsure of the very tiny delay in allowing the lamp the briefest of power just a tiny glimmer before the 10 fuse blew. Maybe the fuse should have been a bit lower rating, I'm not sure.
 

Offline Jwillis

  • Super Contributor
  • ***
  • Posts: 1673
  • Country: ca
Re: Zener diode dissipation for protection.
« Reply #12 on: May 16, 2021, 10:06:51 pm »
Like you I have many things to learn . It's possible that the maximum gate current was exceeded at 12V . That's just a guess . Did it destroy the SCR?
 
The following users thanked this post: davelectronic

Offline davelectronicTopic starter

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: gb
  • Life's too short.
Re: Zener diode dissipation for protection.
« Reply #13 on: May 17, 2021, 01:03:16 am »
Yes the first SCR didn't survive for some reason. Maybe i exceeded one of the maximum ratings. I'm happy with the result though, i think it is somewhat crude with some loose variations ( not tight tolerances )
But if it provides a safe guard against destroying a transceiver, then it was well worth adding to the psu. Although this first version of a crowbar is going to be an add on, as i have no room inside the power supply case. with an added project box with crowbar circuit and a volt amp panel meter it shouldn't look to much out of place. I have four psu projects unfinished, i can factor a crowbar circuit into those, as i'm still at the hardware layout stage and yet to work out the pcb for those projects. Thanks for all the help, much appreciated.     
« Last Edit: May 17, 2021, 01:10:35 am by davelectronic »
 

Offline Jwillis

  • Super Contributor
  • ***
  • Posts: 1673
  • Country: ca
Re: Zener diode dissipation for protection.
« Reply #14 on: May 17, 2021, 03:45:38 am »
I was guessing you are using a Switch mode power supply . If your using a linear supply you need to work out the voltage at the collector of the pass transistors. If a failure occurs in the pass transistors they will fail shorted allowing the voltage at the collector to pass to the out put . Which can be a few volts higher than the out put voltage .  So you need to take that into account when selecting the limiting resistor for the zener and the SCR .
 
The following users thanked this post: davelectronic

Offline davelectronicTopic starter

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: gb
  • Life's too short.
Re: Zener diode dissipation for protection.
« Reply #15 on: May 17, 2021, 02:33:07 pm »
No I'm not using a SMPS, it's a linear power supply with two toroidal transformers, a L7812 voltage regulator and 4 X TIP36C PNP power transistors. This type of homebrew power supply. One is nearly finished, three others are still in progress, be it slow. The taller units are two cases bolted together, transformers in the lower half. The control circuit and filter capacitors are in the top half of the unit. The nearly finished unit is putting out 12.60 Volts. I've tested up to 15 Amps using a clamp meter, voltage drop to 12.35 Volts under that 15 Amp load. I just wanted to protect against just that scenario of a power transistor going short circuit.
 

Offline Jwillis

  • Super Contributor
  • ***
  • Posts: 1673
  • Country: ca
Re: Zener diode dissipation for protection.
« Reply #16 on: May 17, 2021, 08:23:53 pm »
OK . They look good.
The L7812 voltage regulator has a drop out of 2V . This means that for the regulator to output the rated voltage  the input to the regulator must be at least 2V higher than the out put .  So if the pass transistor (TIP36C) were to fail , they would pass at Least 14V to the out put of your power supply . Maybe more depending on what voltage is before the pass transistor and regulator.
I said collector (different configuration)  before ,  but in your case it would be the emitter voltage  of the PNP pass transistors .
In either case , the zener diode limiting resistors should be calculated for that higher voltage . Minimum resistor value: Rmin=(Vs-Vz)/Izmax
Your building these supplies yourself?
 
The following users thanked this post: davelectronic

Offline Tomorokoshi

  • Super Contributor
  • ***
  • Posts: 1212
  • Country: us
Re: Zener diode dissipation for protection.
« Reply #17 on: May 18, 2021, 03:37:07 am »
Let's run some math. I'm estimating some of these.

SCR Crowbar:
AGC 10 fuse: 0.01 Ohms.
Power supply bulk capacitor voltage: 25V
TYN1225 surge current limit: 300 A (which you get to do only once)
TYN1225 resistance at 300 A: 4V / 300A = 0.013 Ohms
Trace resistance: 0.02 ohms

Double check SCR clamping:
(25 - 4) / (0.01 + 0.013 + 0.02) = 488 A

So at least it's on the order of what it takes to damage the part.

TVS Clamp:
AGC 10 fuse: 0.01 Ohms
Power supply bulk capacitor voltage: 25V
TVS clamping voltage: 16 V
Trace resistance: 0.02 ohms
Current through fuse: (25 - 16) / (0.01 + 0.02) = 300 A

This is well past the limit of a 1.5KE15A of 71.7A.

If using a 5KP13A, the maximum current is 237A. While one is not enough, two or three in parallel would load share and have enough capacity.

Fuse Calculations:
AGC 10 fuse: 0.01 Ohms.
Typical I^2t = 486 A^2s
At 300 A the time to open is around (486 A^2s) / (300A ^2) = 5.4 ms

So you get around 5.4 ms of passthrough before the fuse opens. This is on the order of the 8.3 ms sine wave half cycle. However, this is only through the fuse and TVS. It is still clamping to 16V. For the SCR crowbar, you might get the 25V exposed to the load before the entire circuit activates.
 
The following users thanked this post: davelectronic

Offline davelectronicTopic starter

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: gb
  • Life's too short.
Re: Zener diode dissipation for protection.
« Reply #18 on: May 18, 2021, 07:21:00 am »
The rectified filtered voltage is 20.56 Volts, before any transistors or regulator. I've been interested in power supply electronics for years, but early incarnations current was not beyond 5 Amps at 13.50 Volts the kind of power supply in the picture below. Yes I built them from scratch, one has two rewound microwave oven transformers inside, wired in series. The other two have toroidal transformers in them. I wish my maths was as good as my engineering skills. These are long time projects, life events robbed me of a bright future. I get help from my Son with heavy tasks like steel cutting and drilling etc. They've been sitting around for a long time now, it would be great to finish them. I'm not sure that's doable now. But I will give it my best to try. I get some of the maths you've listed there, but there isn't going to be 300 Amps flowing through anything I'm using, so no maximum ratings will be exceeded. I have good head room voltage for the L7812 regulator. I've just remembered, one PSU has a LDR 12 volt regulator. This is because of an overall lower voltage from the two rewound microwave oven transformers with high current secondary windings. 14 AWG from memory. These are wired in series, both primary and secondary windings. I would like to try an LM723 PSU project, but as said, that might not be possible now.
 

Offline Jwillis

  • Super Contributor
  • ***
  • Posts: 1673
  • Country: ca
Re: Zener diode dissipation for protection.
« Reply #19 on: May 18, 2021, 07:50:58 am »
All of them look great . Welll done!  :-+
 
The following users thanked this post: davelectronic


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf