heating up of the MC34063AP IC

[Joakim]

hi, am so greatful for whatever your doing cause its wonderful. well i was trying to come up with a charge regulator for a 12V,7Ahr sealed lead acid battery. i used the mc34063 IC as a step-down together with its associated electronic components as in the systematic lay out in the step-down configuration.

i supplied the regulator with 20VDC and i wanted 14.7VDC out and a current of 0.8A.

it regulated the voltage very well and i think the current too.

however it heat up in 1.5munites just after i had connected the battery on it; and the temperature shot to 210 o F in 4 minutes.

As the temperature goes high, the current from the power supply decreases and the current to the battery also decreases.
i tried cooling the IC by attaching a heat sink on to it but still the temperature shot up to 230 o F. i beg for any solutions and advice on what to do.

 

[Harald Kapp]

Show us your schematic of the circuit. You probably have a short circuit somewhere or a component that is not suited to the task.
Suspects are:
- the inductor (may saturate if not specified for the current)
- the freewheeling diode (check orientation and type)

 

[KrisBlueNZ]

The MC34063 is a very old device. Switching regulators have improved a LOT since it was popular. They use N-channel MOSFETs which have very low ON-resistance, and a capacitor charge pump to generate a positive rail so the MOSFET can be switched very quickly and fully saturated. Losses and heat generation are much smaller than with the bipolar emitter follower architecture of the MC34063.

Linear Technology and National Semiconductor (now owned by Texas Instruments) make ranges of easy-to-use buck switching regulators rated from 0.5A to 5A and more. Have a look at linear.com and ti.com. Linear Technology's parts are more expensive but easier to use, in my experience.

Edit: Harald's suggestions are also good. Check them first. Especially the inductor characteristics.

Can you give us all the details of the inductor? What core material and shape/size? What AL value? How many turns of wire?

 

[Joakim]

Show us your schematic of the circuit. You probably have a short circuit somewhere or a component that is not suited to the task.
Suspects are:
- the inductor (may saturate if not specified for the current)
- the freewheeling diode (check orientation and type)

hi harald. thanks for the respond. am greatful. well the schematic is simillar to the one the link below though at co i used 100uF, the inductor is 15uH, R1=1.2K, R2=13.2K and Rsc=.o15

 

[KrisBlueNZ]

15 microhenrys is very low. The MC34063 isn't designed to run above 100 kHz.

There's no link in your post.

 

[Joakim]

The MC34063 is a very old device. Switching regulators have improved a LOT since it was popular. They use N-channel MOSFETs which have very low ON-resistance, and a capacitor charge pump to generate a positive rail so the MOSFET can be switched very quickly and fully saturated. Losses and heat generation are much smaller than with the bipolar emitter follower architecture of the MC34063.

Linear Technology and National Semiconductor (now owned by Texas Instruments) make ranges of easy-to-use buck switching regulators rated from 0.5A to 5A and more. Have a look at linear.com and ti.com. Linear Technology's parts are more expensive but easier to use, in my experience.

Edit: Harald's suggestions are also good. Check them first. Especially the inductor characteristics.

Can you give us all the details of the inductor? What core material and shape/size? What AL value? How many turns of wire?

well, thanks so much KrisBluesNZ for the advise, that good information but can you do me a favour and send me a list of some of the switching regulators you recommend me to use and can be better for my exercise. thank you very much.

 

[Joakim]

15 microhenrys is very low. The MC34063 isn't designed to run above 100 kHz.

There's no link in your post.

oh really, am sorry well the schematic is . and please can you give me advise on which will be the best components for the circuit. i have just began this and i would like to learn more, if possible you can send me a pdf where i can read. i just used a drawing of this schematic to construct this circuit. otherwise thank you so much for your advise.

 

[KrisBlueNZ]

That diagram is straight out of the data sheet. What value are you using for CT? Are you sticking with 470 pF? That will run the MC34063's oscillator at about 50 kHz.

Why are you using a 15 uH inductor? The data sheet suggests a 220 uH inductor with CT = 470 pF. 15 uH is much too small.

What are the other characteristics of the inductor? Is it an off-the-shelf component? If so, can you link to the data sheet? If not, can you tell me the former style, the core material and AL value, and the number of turns?

In any case you will have to use a much higher inductance, or a more modern IC. Look at the LT1372, it is probably much better suited.

Edit: Also check out the LM25011.

 

[Joakim]

That diagram is straight out of the data sheet. What value are you using for CT? Are you sticking with 470 pF? That will run the MC34063's oscillator at about 50 kHz.

Why are you using a 15 uH inductor? The data sheet suggests a 220 uH inductor with CT = 470 pF. 15 uH is much too small.

What are the other characteristics of the inductor? Is it an off-the-shelf component? If so, can you link to the data sheet? If not, can you tell me the former style, the core material and AL value, and the number of turns?

In any case you will have to use a much higher inductance, or a more modern IC. Look at the LT1372, it is probably much better suited.

Edit: Also check out the LM25011.

thanks for the reply,
well i used the 15uH cause its what i had and i thought trying it out just to see if it could work. for CT i still used the 470pF.

just to ask you, could you give me any directions on how i can cool an inductor. the reason as to why am asking you such a favour is where i am in Uganda its really difficult to find electronic components, the suppliers here only have a few of them. i really would like to learn more of practical electronics but am limited by the fact that i have to make orders to outside manufacturers to import most of the components. here i beg you if you could recommend me to good suppliers of good electronic components cause feck products are too much this side.

thanks for all you am really great full for your time

 

[(*steve*)]

Cooling the inductor is not going to work.

An inductor works like an inductor until the current reaches a certain value (saturation current). It then stops working like an inductor and starts working like a short circuit.

When it's acting like a short circuit, excessive current passes through it, and both it and the rest of the current path (in this case your IC) get very hot.

The cure is to:

1) use an inductor with a higher saturation current (but this probably won't work in your circuit)

2) get an inductor with a higher inductance (and adequate saturation current)

3) reduce the duty cycle considerably

4) increase the frequency considerably.

You have an inductor that is less than 10% of the recommended value. Sure, you can often tweak things, but not often by a factor of 1000%

If you can get more if these inductors, you can try placing several in series. At some point (maybe with 10 or more of them) the circuit will operate more or less normally.

You could also try winding your own inductor, but that will require either a lot of wire, or a core that you know something about.

 

[KrisBlueNZ]

Steve has explained the situation thoroughly, as usual.

You can increase the operating frequency by reducing CT. 470 pF gives an operating frequency of about 33 kHz but the MC34063 is specified to operate at up to 100 kHz. The CT value for 100 kHz operation is about 180 pF but this value comes from a "typical performance" graph so it is not very exact. You should check with a frequency counter (on pin 2) and adjust CT slightly if necessary.

A combination of increasing the frequency to 100 kHz and connecting FIVE 15 uH inductors in series should give behaviour equivalent to the data sheet diagram. You can try using fewer inductors. You could also try increasing the frequency beyond 100 kHz but you can't really be sure what will happen!

 

[Joakim]

Cooling the inductor is not going to work.

An inductor works like an inductor until the current reaches a certain value (saturation current). It then stops working like an inductor and starts working like a short circuit.

When it's acting like a short circuit, excessive current passes through it, and both it and the rest of the current path (in this case your IC) get very hot.

The cure is to:

1) use an inductor with a higher saturation current (but this probably won't work in your circuit)

2) get an inductor with a higher inductance (and adequate saturation current)

3) reduce the duty cycle considerably

4) increase the frequency considerably.

You have an inductor that is less than 10% of the recommended value. Sure, you can often tweak things, but not often by a factor of 1000%

If you can get more if these inductors, you can try placing several in series. At some point (maybe with 10 or more of them) the circuit will operate more or less normally.

You could also try winding your own inductor, but that will require either a lot of wire, or a core that you know something about.

cool steve, thanks for the advise, how ever still my problem stands that i can afford getting the right inductors in ma country. in most cases i have to order from out (importing) but i really find it hard getting the right supplier.


I like that idea of making ma own inductor though i have no idea on what to do . i beg if you could give me some crue on how to cool them that would be cool. you can even send may be a pdf to ma email address ([email protected]) i will be great full.

thank you.