Would you please have a look at my calculations to see if I’m on the right track as I

Would you please have a look at my calculations to see if I’m on the right track

Hello.


Would you please have a look at my calculations to see if I’m on the right track as I don’t really know which values to use in the datasheet?

I want to drive an h-bridge with TIP 140 power transistors, View attachment TIP140 .pdf.

So the idea is to turn the transistor on by applying the correct base current I_b while I know what my collector current I_c is and I’m using 5V from a pwm as base voltage V_b.

I_c = 5A

So

I_b = I_c/hfe

hfe is found in the datasheet to be 1000 at I_c = 5A.

I_b = 5mA

And then

R_b = (V_b - V_be )/ I_b

Vbe is taken as 3V from the datasheet.

R_b = (5V – 3V )/ 5mA

R_b = 400ohm

Is this how it should be done?
If I want this transistor to turn on and off as fast as possible or to make it move across the active region to saturation as fast as possible should in not just make Ib swing from 40mA to -40mA to keep ton and toff to a minimum like it is shown in the datasheet?

Yes, you've gone about it the right way for the most part.

Question: Is your spec for biasing the transistor so it passes exactly 5A to the load? Or do you wish to drive the transistor at its minimum ON resistance?

Because normally you would send sufficient current to the base until the transistor is down to its minimum ON resistance. This ensures it will not heat up as much, and won't waste much power.

In any case, whether you're using the transistor as a switch, or to deliver a specific current, then 400 ohms is a reasonable bias resistance.

In practice you'll put a pot and maybe 220 ohm resistor in series. Then you can increase current into the base without fear of frying it. When you're happy, substitute a single resistor.

The NPN type will go in the low side of your H bridge.

PNP type should go in the high side. Adapt your pulse polarity cycle accordingly.

If I want this transistor to turn on and off as fast as possible or to make it move across the active region to saturation as fast as possible should in not just make Ib swing from 40mA to -40mA to keep ton and toff to a minimum like it is shown in the datasheet?

Click to expand...

A healthy transistor permits current to flow only one way through the base.

Current may go the other way but only if you apply sufficient voltage to push it into zener operation. This is an atypical way to operate a transistor.

I'm not sure why the datasheet quotes a -40mA figure among data for an NPN.

Your circuit should work well enough by alternating the bias signal between zero and 5V.

Question: Is your spec for biasing the transistor so it passes exactly 5A to the load? Or do you wish to drive the transistor at its minimum ON resistance?

Click to expand...

I actually want the transistor to go completely on as fast as possible and off in the same manner so it spends the least amount of time in the active region where the power dissipation in the transistor would be at its highest. I believe this would keep my circuit optimized as that is what I am going for.

I’m sure this answered your question with regards to what I want to achieve with regards to the biasing of the transistor.


Should I not be more concerned with the effect the capacitor Cbe will have on the transient response of the transistor as this is what would keep it in the active region longer?

Maybe this is where the 40mA and -40mA come in play with regards to biasing of the transistor as a larger base current would charge capacitor Cbe up faster and the -40mA will discharge Cbe faster without any current flowing through the emitter to the base as you mentioned and this will then result in the Ton and Toff of 0.9ns and 4ns respectively as indicated in the datasheet.

komplex said:

I actually want the transistor to go completely on as fast as possible and off in the same manner so it spends the least amount of time in the active region where the power dissipation in the transistor would be at its highest. I believe this would keep my circuit optimized as that is what I am going for.

Should I not be more concerned with the effect the capacitor Cbe will have on the transient response of the transistor as this is what would keep it in the active region longer?

Click to expand...

Yes, you have targeted a topmost concern when switching high current at hi frequency.

Incidentally... Are you switching a coil with your H bridge? It may generate hi-voltage spikes. The faster your transitions, the more severe the spikes. That (or some other problem) may draw most of your attention.

Maybe this is where the 40mA and -40mA come in play with regards to biasing of the transistor as a larger base current would charge capacitor Cbe up faster and the -40mA will discharge Cbe faster without any current flowing through the emitter to the base as you mentioned and this will then result in the Ton and Toff of 0.9ns and 4ns respectively as indicated in the datasheet.

Click to expand...

Yes, in theory. But I don't recall that they started manufacturing transistors so it's normal to do that.

As to what negative voltage will reverse-bias the TIP device... it doesn't seem to be specified in the datasheet. In fact it's rarely mentioned anywhere because it's not a normal way to use a transistor. I've seen figures of 7 and 10 V.

And notice that pulling current both IN and OUT of the base will require you to create a drive source that delivers +5 V one moment, and the next moment a negative voltage which is more negative than the TIP ground leg. Does your project already have a negative supply rail?

You can experiment of course. You may discover a better way to use a transistor. It's only fair to let you know it may also result in frying a device or two.

And now, thinking about it further... I believe the -40mA quote is for that amount of DECLINE in bias current. From more positive to less positive. So the turn-off time is 4 uS when you drop from 40mA to 0 bias. Or from 50 to 10mA. Or from 60 to 20mA.

And the turn-on time is .9 uS when you raise it from 0 to +40mA bias. Or 10 to 50mA. Or 20 to 60mA. Etc.

Firstly I’d like to thank you for your reply, I really do enjoy bouncing ideas and such things as it does help me understand things more clearly.

Does your project already have a negative supply rail?

Click to expand...

I do not have a negative rail, so I’ll just abandon the whole idea of it.

And now, thinking about it further... I believe the -40mA quote is for that amount of DECLINE in bias current. From more positive to less positive. So the turn-off time is 4 uS when you drop from 40mA to 0 bias. Or from 50 to 10mA. Or from 60 to 20mA.

And the turn-on time is .9 uS when you raise it from 0 to +40mA bias. Or 10 to 50mA. Or 20 to 60mA. Etc.

Click to expand...

That does make a lot more sense. I found another datasheet for the same device but different manufacturer that shows it clearer and also support your concept of it.



I was told that the H-bridge should be able to go forward and reveres with speed control and that if the circuit has a duty cycle of 50% it should have holding torque on the motor as it will be attempting to go forward and reveres in one period, I have designed my circuit as such but when I tried to test it I blew the 10A fuse in my multimenter.

This leads me to believe that having the motor stand still while it is attempting to move but is not doing so is a short circuit?