So, I am working on a project where I am running about a 300mA load through a transistor. I have tried two different transistors, one rated at 500mA(2N-4401-218), and the other rated at 1A(C1384-031). They both burn up, and I am wondering if I have misunderstood the Amp ratings, or if I just need to heat-sink these somehow. ? ? ?

Here is the schematic: The 18 ohm and 15 ohm loads represent incandescent lights. They tend to be pulling about 300mA each



I also checked my power dissipation, the 2N4401 is rated at 630mW I checked my power dissipation and calculated it to be 301mW, (Vce*Ice+Vbe*Ibe) = (1.170*0.780+0.256*0.00126) = 301mW.

 

Post your .asc file if you can.

 

Have you allowed for the duty cycle?
For Q6, LTspice says :-
peak current =242mA,
average current = 78mA,
peak power = 476mW,
average power = 147mW.

If your transistors are frying, those figures suggest you have a wiring error or wrong component values.
Check the pinouts of the transistors.

 

This looks almost like an astable multivibrator (flasher), except that there are 3 stages.

I would try replacing the incandescent lamps in the physical circuit with 15 or 18-ohm resistors, as applicable, and see if the transistors still burn. If they suddenly survive, then it could be that the lamps have less resistance in real life then you expected, especially when run at a duty cycle less than 100%. There is a current surge when a cold lamp is turned on, and maybe the transistors are seeing all current surges in operation.

 

Please spread out the components on your schematic so it is readable.

First, power dissipation is in mW, not mA. Second, check your math. The total power dissipation is over 0.9 W.

Third, you are driving the loads with emitter followers rather than saturated switches. This increases greatly the power dissipation in the controlling device.

ak

 

This looks almost like an astable multivibrator (flasher), except that there are 3 stages.

I would try replacing the incandescent lamps in the physical circuit with 15 or 18-ohm resistors, as applicable, and see if the transistors still burn. If they suddenly survive, then it could be that the lamps have less resistance in real life then you expected, especially when run at a duty cycle less than 100%. There is a current surge when a cold lamp is turned on, and maybe the transistors are seeing all current surges in operation.

That's a good thought. Once I find enough small resistors to try that, I will. I could only find enough little resistors to give me 45 ohms; those only pulled 100mA. But I have also tried lighting up one channel at a time, without them chasing. The transistors really heat up when the single channel stays on. This tells me that it is not a cold-start surge. The only time that I have had any transistors burn up was while I had them stopped on one channel. While they chase, they have gotten pretty warm, but I don't let them run long enough to burn out.

 

are Q1 Q2 and Q3 burning up or Q4, Q5, Q6 burning? I always use the transistors rated more than the circuit requirement.

 

So, I am working on a project where I am running about a 300mA load through a transistor. I have tried two different transistors, one rated at 500mA(2N-4401-218), and the other rated at 1A(C1384-031). They both burn up, and I am wondering if I have misunderstood the Amp ratings, or if I just need to heat-sink these somehow. ? ? ?

Here is the schematic: The 18 ohm and 15 ohm loads represent incandescent lights. They tend to be pulling about 300mA each

View attachment 93866

I also checked my power dissipation, the 2N4401 is rated at 630mW I checked my power dissipation and calculated it to be 301mW, (Vce*Ice+Vbe*Ibe) = (1.170*0.780+0.256*0.00126) = 301mW.

Filament bulbs have a PTC characteristic - that is the cold resistance is much lower than the value calculated from normal operating current and voltage. Its one possibility that the transistors can't handle the cold current surge at turn on.

 

Post your .asc file if you can.

 

are Q1 Q2 and Q3 burning up or Q4, Q5, Q6 burning? I always use the transistors rated more than the circuit requirement.

Q4, Q5 and Q6. Their maximum ratings are about double the circuit requirement by my calculations.

 

I have been letting my circuit run for a little while. The transistors get fairly warm, but they haven't gotten too hot yet. I plan to enclose the circuit in a small box though; they might overheat then.

This Astable Multivibrator circuit is always touchy, I have a hard time finding resistor and capacitor values that work well. One combination worked well until I changed out the small transistors for the slightly larger ones,(Q4,Q5,Q6); I am assuming that the larger ones require a higher base current. To get the circuit started now, I have to pull out a capacitor and stick it back in. I am using the same values that are in the image that I posted.

 

I have been letting my circuit run for a little while. The transistors get fairly warm, but they haven't gotten too hot yet. I plan to enclose the circuit in a small box though; they might overheat then.

This Astable Multivibrator circuit is always touchy, I have a hard time finding resistor and capacitor values that work well. One combination worked well until I changed out the small transistors for the slightly larger ones,(Q4,Q5,Q6); I am assuming that the larger ones require a higher base current. To get the circuit started now, I have to pull out a capacitor and stick it back in. I am using the same values that are in the image that I posted.

There used to be Japanese TO92 transistors potted into an aluminium "P" clip, they had higher dissipation capability than regular TO92, but were evidently intended for bolting to a heatsink.

For driving lamps; I'd probably pick something in at least TO126 package. There is a more recent package type that was originally SMD, but was also made available in through hole format - I think it was something like TO251.

 

There used to be Japanese TO92 transistors potted into an aluminium "P" clip, they had higher dissipation capability than regular TO92, but were evidently intended for bolting to a heatsink.

For driving lamps; I'd probably pick something in at least TO126 package. There is a more recent package type that was originally SMD, but was also made available in through hole format - I think it was something like TO251.

Cool, thanks. I found the TO251 on ebay; it is 1W dissipation like the ones that I am using already, but it can handle a much higher collector current. I don't know if that will change things or not, but it is meant to be used for driving lamps.

 

See transistor http://www.diodes.com/_files/datasheets/ZTX690B.pdf

http://www.digikey.com/product-detail/en/ZTX690B/ZTX690B-ND/92520

 

Cool, thanks. I found the TO126 on ebay; it is 1W dissipation like the ones that I am using already, but it can handle a much higher collector current. I don't know if that will change things or not, but it is meant to be used for driving lamps.

I think for house lightbulbs; the surge current is something like 12x the normal running current, no idea what it is for low voltage but it may well be "stiffer".

The transistor has to cope with that turn on surge even though the steady running current is much less. On the bright side - the transistor that can handle it, probably won't need a heatsink as long as its properly saturated.

In the past, I've seen projects that use a low value resistor to pre-heat the filament. Just a dim glow you can barely see if you cup your hand round it, but it gives the transistor a head start and eliminating most of the turn on surge makes the bulb last longer - but it wastes a lot of energy.

 

How about MOSFETs? I found these transistors on ebay. They have a high power dissipation for their size, and MOSFETs shouldn't dissipate as much power as the bipolar transistors do.

Will these work in a typical Astable Multivibrator circuit though? I'm not very familiar with the differences in transistors. I know that the assumed gate current in a MOSFET is 0 though, and that will change some things.

 

MOSFETs should work really well for driving my lamps though; they will not take current away from my main circuit. Right now, I have to place resistors between my circuit and the base of the driving transistors so that the capacitors will charge and get the circuit moving.

 

How about MOSFETs? I found these transistors on ebay. They have a high power dissipation for their size, and MOSFETs shouldn't dissipate as much power as the bipolar transistors do.

Will these work in a typical Astable Multivibrator circuit though? I'm not very familiar with the differences in transistors. I know that the assumed gate current in a MOSFET is 0 though, and that will change some things.

The IGBT came into being because early MOSFETs didn't perform well on RDSon and couldn't match the VCEsat of transistors at the time - The IGBT is basically a hybrid Sziklai pair by hanging a PNP emitter follower on the drain of an N MOSFET.

Modern MOSFETs have very impressive RDSon figures - but you have to satisfy VGSthr to make that actually happen. A typical power MOSFET may need as much as 10V, logic level types may work with VGS as low as 4V.

 

I am an electronics hobbyist and I am not a representative for which I am endorsing. I was purchasing parts through e-prey where in a purchase I found a site where i earn $12 mil a week, i hate those stupid ads, but I found TAYDA Electronics and hardly ever buy any parts from e-prey at all.
Check them out everyone and maybe you will be amazed at the prices like I was. Digi-key is another site but $$.
By the way could I ask you what type of application are you tackling? I build all kinds of wacky things to save money or just play with.