Sign in to follow this  
Followers 0
Taggsladder

Learning transistors and I have a problem :)

6 posts in this topic

Hello

 

I am working on a curcuit that need to switch two loads max 200mA. One load gets feed ground and the other load gets feed positive. VCC will be in a span of 6-30VDC. The curciut works as intended when VCC is 6VDC, but at the max span VCC 30VDC all (Q1, Q2, Q4) transistors and the FET breaks when the pin from the MCU goes HIGH. This will happen even if the connector is unconnected.

 

post-36838-0-19328600-1443614451_thumb.jpg

 

Q1, Q2

http://www.farnell.com/datasheets/699988.pdf

 

post-36838-0-36281300-1443614445_thumb.png

 

 

 

 

Q4

http://www.farnell.com/datasheets/1885683.pdf

 

post-36838-0-55051100-1443614448_thumb.png

 

 

From what stupid me can read from the datasheet Q1 and Q2 should be able to handle 45V between collector-emitter. But I think that I can understand why Q4 fails becouse it can handle 20V max between gate-source. Am I right? :ph34r:

 

Thanks

 

Best regards,

Andreas

Share this post


Link to post
Share on other sites

Q4 is going to see greater than 20Vgs, and may fail. Generally not a lot of leeway in these ratings. I wouldn't even count on a long life if you are running at just below 20V, as even small transients beyond 20V can damage the FET.

 

As for the two BJT's, I don't see any particular issue with Q2 until the gate of Q4 fails, as there shouldn't be dissipation greater than about 225mW while switching for Q2 at 30V supply (15mA*15V), and a few mW when full on. That shouldn't be sufficient to lead to heating issues. Once the gate fails for Q4, though, there is indeterminate current limiting for the collector of Q2, and likely Q2 will fail.

 

I would also put a 100K to ground from the base, though, to prevent leakage from turning it on, as the LED in the base line keeps the output pin from pulling it hard off.

 

I can say nothing about Q1 failing without knowing the diode you used. What is its reverse breakdown?

 

 

Once you are switching a load, things get more involved, as the dissipation in Q1 can be significant, especially if Ib is deficient. You need about 10mA Ib to hold the transistor in saturation at 200mA Ic, and I don't think you will have that here, which means you may end up with significant dissipation under load, if you are running at 200mA. You may also have dissipation issues with Q4 under load, but I am too lazy to look up the full data sheet right now. Things may be OK-- unless there is a major issue looking at the data sheet, I would go with this as a start for 100 to 200mA-- but I wouldn't put a product in the field without a lot of testing, as you are closer to the bounds than I might be comfortable with based solely on a paper design.

dubnet and Taggsladder like this

Share this post


Link to post
Share on other sites

Thanks for the reply @@enl!

 

Ok, I understand. The reverse breakdown of D5 is 200V.

 

D5

http://www.farnell.com/datasheets/162697.pdf

 

Though I think that I will go with the original plan of just using a DPDT signal relay switched by a transistor, find it is easier when you know exactly whats gonna be switched. Would have been neat to use them magic FET's but clearly I have some learning todo :)

 

This forum is such a great place, thanks again!

 

Kind regards,

Andreas

Share this post


Link to post
Share on other sites

Depending on the application, you might try a latching relay instead.

 

In a low power ap, you can't beat them, but there are a few drawbacks, such as they do not drop out on power loss like a standard relay. Even in a non-low power ap, not needing to power the coil continuously is a nice feature. If you haven't seen a latching relay, among other places you see them is in electronic thermostats.

Taggsladder likes this

Share this post


Link to post
Share on other sites
Hello,

 

Forget relay, just whack an 1k resistor between the G of the Q4 and C of the Q2 and put a 10V zener across that 10k resistor and the Vgs of the Q4 is OK. Use a N-FET instead of the Q1 & D5 and you are ready to go.

 

In fact, I would replace the Q4 with an N-FET too, just as a good practice. If MOSFETS cost half a penny each, why we still use BJTs? You can fly to the Jupiter just using the base currents of all those misused BJTs.

 

Cheers,

 

-=solipso=-

Taggsladder likes this

Share this post


Link to post
Share on other sites

Depending on the application, you might try a latching relay instead.

 

In a low power ap, you can't beat them, but there are a few drawbacks, such as they do not drop out on power loss like a standard relay. Even in a non-low power ap, not needing to power the coil continuously is a nice feature. If you haven't seen a latching relay, among other places you see them is in electronic thermostats.

 

Great id

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0