1. Home
  2. Electronics Design
  3. How to blow up an LM395

How to blow up an LM395

formatting link

I use these as "bullet proof" output pass elements. One is used to drive a 50 ohm heater. (~30V p-p PWM) I had a few reports that the circuit would not turn off, it was always heating. I had one user ship the board back. Indeed the LM395 was toast. I tried everything to destroy another one, Short circuit the load, change the load... I left the LM395 thermally cycling for hours.. but it still worked afterwards.

So the heater wires are bundled with the thermal couple (sensor wires) and plug into the back of the electronics, (banana plugs for heater.) In the same piece of apparatus there is a big (100 Gauss) Set of Helmholtz coils. (R ~10 ohms) These also have banana plugs!

If I plugged the coils into the heater output, then "blamo", It took out the LM395... the poor puppy could never turn off. A diode across the output now allows the LM395 to drive big the big inductor and survive. :^) I think we'll add a warning label on the coil too.

No real question here... just a story from the trenches.

George H.

Reply to
George Herold
Loading thread data ...

According to JT, the LM395 is just a metallization option of the LM317, and so has pretty much the same strengths and weaknesses. In particular, shoving much current backwards into the output will fry it. I suspect it was oscillating, and the back EMF from the coil killed it.

The other two reliable ways to kill one are input overvoltage and big-time reverse current on the adjust (base) pin.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net
Reply to
Phil Hobbs

Yep...

Current _up_ thru the substrate diode is a guaranteed killer. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142     Skype: skypeanalog  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
I love to cook with wine.     Sometimes I even put it in the food.
Reply to
Jim Thompson

Well I didn't spend a lot of time diagnosing the failure, but I couldn't see any sign of oscillations. (I was expecting to see something turning on and off.) The output went to the 1/2 power point (~15V), Sat there for ~15 minutes.. until it cooked itself and then jumped to full on.

George H.

Reply to
George Herold

Weird. That shouldn't kill it. Of course its slew rate is the pits, so a low level oscillation might not be easily visible.

Does it die with just a 10-ohm resistor?

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net
Reply to
Phil Hobbs

Hi Phil, Well I think it was trying to oscillate internally... once I added a diode across the output that's exactly what it did. (At a few milli second time scale.) It is a relatively large air coil. Two ~4" radius coils ~500 turns per side. Once current was flowing it just couldn't stop itself... or something like that.

With the diode across it, or any other resistive load it lasts forever.. Well at least several hours.

George H.

Reply to
George Herold

Well, that's one for the files, for sure. Glad you got it figured out.

Cheers

Phil Hpbbs

Reply to
Phil Hobbs

George, Were you driving the load from the 'collector' or 'emitter'? (That is, was it input or output inductance that'll kill a 317/395?)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net
Reply to
Phil Hobbs

The emitter. An opamp driving it as a voltage follower. Collector to +30V, emitter to grounded load.

George H.

Reply to
George Herold

Its a common problem driving inductive loads, we have a 40 amp stud mounted diodes on a magnetic amplifiler to capture unexpected out of control inductive collapes.. Otherwise, it takes out a frew HV transistors in quasi configuration.

Jamie

Reply to
M Philbrook

I see. So if it wasn't oscillating, probably it just hit the thermal cutout, tried to turn off, and then the load dragged the 'emitter' 30V below ground and blew it up from overvoltage. Hence the diode to ground fixed it, gotcha.

I try to make my protos idiot-resistant as well, not just my finished designs. After a day patiently bashing away at some baulky apparatus, by 5 PM I can do some pretty silly things, and having banana plugs everywhere does _not_ help.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net
Reply to
Phil Hobbs

Yeah, the fault does point back to me. And this just serves as a reminder, that people are going to plug things together if it is at all possible. My job is to make sure nothing breaks when that happens...

George H.

Reply to
George Herold

Students, squaddies, and bureaucrats.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net
Reply to
Phil Hobbs

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.