Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Problem with ST MC34063C DC to DC converter

Status
Not open for further replies.
B

boylesg

Advanced Member level 4
Joined
Jul 15, 2012
Messages
1,023
Helped
5
Reputation
10
Reaction score
6
Trophy points
1,318
Location
Epping, Victoria, Australia
Activity points
11,697
I have wired up the following circuit:


The image is reversed such that the pinout matches that of the actual chip.
And the component values in this circuit are not what I have actually used.

My own calulcations are as follows:

VOUT (V): 17.500000
IOUT (mA): 500.000000
VF DIODE (V): 0.750000
VSAT (V): 0.700000
FMIN (kHz) : 80.000000
VRIPPLE (V): 0.010000
tON / tOFF: 0.553097
(tON + tOFF): 0.000013
tOFF: 0.000008
tON: 0.000004
CT (pF): 178.062678
IPK (mA): 1553.097345
RSC (Ω): 1553.097345
CO(ųF): 2003.205128
LMIN(ųH): 32.388637
R1 (Ω): 1200
R2 (Ω): 15600

I have used a 100uH inductor, 0.22R for Rsc, 2200uF for Co and 201/200pF for Ct.

Here is the actual circuit - done with the short circuits one solderless kit:


12V in and I get between 11V and 12V out. The calculations say it should be about 18V out.

12V input is on the right hand side.
Output is on the left hand side and I have added a 10k wire wound resistor as a load across it.

Why wont this f'ing thing work? I can't understand what I could be doing wrong as there is not that much to this circuit.

Strange thing is that I wired this up on my bread board and I did get it working with 18V out.

But I decided to wire it up using my son's Short Circuits kit because that corresponds directly to the prototype boards I am using. But apparently I have made exactly the same mistake with this as I did with the prototype board I soldered up. But I am damned if I can see what the mistake is.

Can anyone spot it or predict what it is likely to be?
 

Hi, Boylesg.

>IPK (mA): 1553.097345 < - is too big chip will be powerfully warmed and will not provide reliable work.
The breadboard not best way to make sure in stability of the high-frequency schemes.
Try to change R1 >> 150 Ohm, R2 >> 2K, Ct >> 470pF, inductor better change too on more solid.

12V in and I get between 11V and 12V out
Click to expand...
But probably scheme is wrong collected and simply does not work. On photographies it is difficult to see details.
 
Last edited:

DeepOne said:
Hi, Boylesg.

>IPK (mA): 1553.097345 < - is too big chip will be powerfully warmed and will not provide reliable work.
.
Click to expand...
Thanks Deepone. What should IPK ideally be then?

I presume that this is related to the ISW in the maximum ratings which is 1.5A

DeepOne said:
Try to change R1 >> 150 Ohm, R2 >> 2K, Ct >> 470pF, inductor better change too on more solid.
Click to expand...

Are you suggesting that I am trying to run the chip at a frequency that is too high to be reliable? If so what is the most often used frequency range for this chip?
The inductor is from the power supply section of a tv circuit board and has quite coarse guage wire. I would have thought that it could handle a few amps.

DeepOne said:
But probably scheme is wrong collected and simply does not work. On photographies it is difficult to see details.
Click to expand...
See that is the problem. There is not much to get wrong with this circuit and I have been over it in conjunction with the shematic a few times and can't see any mistakes. So there must be something else wrong.
 

"12V in and I get between 11V and 12V out" - this signifies that or no generations or output transistor is not connected. Can be a faulty 34063.
What should IPK ideally be then
Click to expand...
I think no more than 1A.
Are you suggesting that I am trying to run the chip at a frequency that is too high to be reliable?
Click to expand...
With increase the frequency requirements to arrangement of the design are tightened.
 
Last edited:

KonstanU said:
Typical frequency is 30Khz. Change Ct = 1n. (At least try to see result)
Click to expand...

Changed Ct to 1.5nF.

I spotted another mistake too. I had the other end of R2 connected to GND instead of to Co. I think that is lesson in walking away when you get frustrated and coming back with a fresh set of eyes.

But 30kHz is quite a bit less than the 80kHz I was using.

Without going into to much detail, can you explain deepone's statement about higher frequency requiring tightening of the arrangement
 

DeepOne is right. The problem of your board is that it was done not for high speed process. It means that your board has a long wires (ground path, signal path and so one) and has a large parasitic inductions, that affect on frequency.
From my experience I need notice that MC34063 doesn't work good on high frequency. In my applications I used 30-50Khz frequency not higher.
 

A layout like the one you're using has lots of stray inductance which acts like high impedances at high frequencies. There will also be lots of high frequency interference between nodes due to mutual inductance. The combination of high impedance leads and interference will typically cause poor behavior of the controller, forcing you to use a lower switching frequency. Using a good layout on a PCB is usually necessary to get low parasitic inductances.
 

KonstanU said:
DeepOne is right. The problem of your board is that it was done not for high speed process. It means that your board has a long wires (ground path, signal path and so one) and has a large parasitic inductions, that affect on frequency.
From my experience I need notice that MC34063 doesn't work good on high frequency. In my applications I used 30-50Khz frequency not higher.
Click to expand...

So I take it the parasitic inductions would make for poor voltage and current output stability?

Any way the changes you suggested and my correction of the additional mistake I found means that it works now so thanks.

12V in and about 28V out.

It is purely the voltage divider at pin 5 that sets the output voltage isn't it?

Would it be advisable to plug the output of this into an LM317 based voltage regulator so that I can adjust it easily between 12V and 28V?

Or would the LM317 cause me problems in combination with the MC34063 circuit?
 

Why regulate a high voltage then drop it down again?
Wire a potentiometer to pin 5 and it will let you adjust the voltage out of the MC34063.

Brian.
 

betwixt said:
Why regulate a high voltage then drop it down again?
Wire a potentiometer to pin 5 and it will let you adjust the voltage out of the MC34063.

Brian.
Click to expand...

Was thinking that the voltage regulator might give me more robust voltage regulation an additional output filter given the output voltage rippling.

What if I was to place the pot as you suggest and adjust the voltage to 19V, or there abouts, and then pass it through an LM317 rather than adding an additional filter to the output as suggested in the datasheet.

Would there be any advantage in doing this or would I be no better off than just adding the additional output filter?
 
Last edited:

Using a second regulator will seriously reduce efficiency because you have to generate too much voltage before the LM317 and lose it as heat in both regulators to drop it down again. The improvement in ripple, if any, would be nefligible and unless you were very careful with wiring it could even make it worse. Your best bet is to add an additional filter if necessary and control the voltage at source in the MC34063. Remember that at the relatively high ripple rate you don't need such bulky filter components than if it were a 50Hz/60Hz non-switching regulator. A simple LC circuit would suffice.

Brian.
 

Generally the ripple from SMPS can be reduced quite a bit with proper layout and additional LC filtering stages on the output. I would only resort to linear regulation if you truly need extremely low ripple (like <1mV). And even then you'd probably need something better than a LM317, with better regulation at high frequencies.
 

betwixt said:
Using a second regulator will seriously reduce efficiency because you have to generate too much voltage before the LM317 and lose it as heat in both regulators to drop it down again. The improvement in ripple, if any, would be nefligible and unless you were very careful with wiring it could even make it worse. Your best bet is to add an additional filter if necessary and control the voltage at source in the MC34063. Remember that at the relatively high ripple rate you don't need such bulky filter components than if it were a 50Hz/60Hz non-switching regulator. A simple LC circuit would suffice.

Brian.
Click to expand...

Fair enough.

- - - Updated - - -

hate said:
You may experiment with this calculator for MC34063 here for different settings:
**broken link removed**

Also this:
http://www.nomad.ee/micros/mc34063a/
Click to expand...

I ended up making my own calculator in a spreadsheet using the formulas in the datasheet. I checked the values I was getting against a similar spreadsheet calculator that I found.
 

Alright. That's it. I am done with MC34063A.

The second circuit I have solderd up and it doesn't work in the same way as the prototype.

Obviously these things are just to difficult for an amateur to get working reliably.

Think I will stick with the voltage regulators and move on.

Obviously the only way I am going to learn properly how to use this f'ing things is to join the local radio club and get one of the pros to show me how to do it. Then perhaps I will be able to see what I am doing incorrectly.
 
Last edited:

Don't give up!

If it helps, I'm in my late 50's and in my whole life I've never had any tuition or training in electronics and only had one day training in software. The morning was a crash course in 'C' because I needed it for the afternoon crash course on Unix! Needless to say, by the end of the day my head was about ot explode. I'm just an amateur experimenter like you but from what I taught myself I've been head-hunted to do work for several multi-national electronic giants and worked on high level projects around the World (and further away).

I would say you have shown a healthy curiosity over how things work but perhaps bitten off a bit too much to digest in a short time. Switch mode regulators can be intimidating and frustrating when they fail to do what you expect but don't give up, just look very closely to see what's gone wrong. You learn more by finding out why things have failed than by reading the manufacturers hype about how good their products are. Look at it this way - if someone had made the perfect SMPS IC, why are there hundreds of different types on the market. They all have problems of some sort and part of the 'fun' is finding ways to work around them and design defensively to avoid the pitfalls in the future.

With experience, your frustraton level will diminish and you will feel more confident about using more advanced components and applications. You will eventually stop hating data sheets and start hating the people who change their design requirements just after you sent thier PCB to the fabricator! Those of us who have been around a while will know exactly what I mean.

Brian.
 

Circuit works without problems like as sound and unstable output voltage.
 

betwixt said:
Don't give up!

If it helps, I'm in my late 50's and in my whole life I've never had any tuition or training in electronics and only had one day training in software. The morning was a crash course in 'C' because I needed it for the afternoon crash course on Unix! Needless to say, by the end of the day my head was about ot explode. I'm just an amateur experimenter like you but from what I taught myself I've been head-hunted to do work for several multi-national electronic giants and worked on high level projects around the World (and further away).

I would say you have shown a healthy curiosity over how things work but perhaps bitten off a bit too much to digest in a short time. Switch mode regulators can be intimidating and frustrating when they fail to do what you expect but don't give up, just look very closely to see what's gone wrong. You learn more by finding out why things have failed than by reading the manufacturers hype about how good their products are. Look at it this way - if someone had made the perfect SMPS IC, why are there hundreds of different types on the market. They all have problems of some sort and part of the 'fun' is finding ways to work around them and design defensively to avoid the pitfalls in the future.

With experience, your frustraton level will diminish and you will feel more confident about using more advanced components and applications. You will eventually stop hating data sheets and start hating the people who change their design requirements just after you sent thier PCB to the fabricator! Those of us who have been around a while will know exactly what I mean.

Brian.
Click to expand...

Obviously there is something fundamentally wrong with the way I am translating my prototype circuit to a soldered circuit and I am clearly not going to solve it without in person help from some one who has done it before.

I am intending to go to a few meetings at the local electronics/radio club since I am enjoying dabling in this. I will ask for help with it there and see what happens.

But in the mean time I have wasted enough boards on this and I am over it for the time being.

Time to get back to the sound card oscilloscope and my jacobs ladder versions.

My 24VAC plugpack / half bridge rectifier/ voltage regulator circuits give me my 18V although I might have to put a small value pot on the adjustment so that I have finer control over the output voltage. It has been a little difficult to get it just right with fixed value resistors and the voltage out varies a bit with the load.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top