Question about a DC-DC convertor circuit. (MC34063A)

[Face]

Hi,
For something I am working on, I need to have ~40VDC, where I normally
have 12VDC. I'm planning on getting this:
http://www.st.com/stonline/books/pdf/docs/5257.pdf . On page 6 of
that datasheet, it shows a circuit diagram which makes use of the
MC34063A, that outputs 28V with a 12Vin. I figured (using the formula
in the notes for calculations on page 9), that to get 40V, changing
the resistors to R1=2.7kohm & R2=82kohm should work. Now, should I
change anything else, or should that be fine?

I was thinking though, that I needn't have some of those components (I
could quite easily be wrong), because I am just using this in a power
line, so I don't care about frequency or output ripple at all. Is
this correct? I found this site to calculate the values to use for
components to use with this chip (how lucky!) at
http://www.nomad.ee/micros/mc34063a/index.shtml . Now, if I leave the
fields blank for Vripple, and Fmin (12Vin, 40Vout, 175Iout), I get the
following:

Ct=NaN pF
Ipk=1254 mA
Rsc=0.239 Ohm
Lmin=NaN uH
Co=NaN uF
R=180 Ohm
R1=2k R2=62k (40V)

I don't know what NaN means...but is it implying that I can leave
those parts out? Also, it calculated 62k and 2k for resistor values,
which give exactly 40V (as opposed to the 39.2V of my configuration),
so it definitely beats me ;P.

Thank you very much,

Jacob Evans

--Oh, and so nobody gets mad at me, disclaimer:
I apologize if I have gotten any terms wrong..I just don't know them
that well, as I am relatively new to this kind of thing. If I said
anything wrong enough to make it unintelligible, let me know, and I'll
clearify. If I got anything wrong that you'd like to correct me on,
feel free to do so, it would be beneficial for me . Also, no, I am
not doing this for any kind of assignment, and want help...I'm just
doing it for a project for myself.

 

[Roger Hamlett]

Hi,
For something I am working on, I need to have ~40VDC, where I normally
have 12VDC. I'm planning on getting this:
http://www.st.com/stonline/books/pdf/docs/5257.pdf . On page 6 of
that datasheet, it shows a circuit diagram which makes use of the
MC34063A, that outputs 28V with a 12Vin. I figured (using the formula
in the notes for calculations on page 9), that to get 40V, changing
the resistors to R1=2.7kohm & R2=82kohm should work. Now, should I
change anything else, or should that be fine?

I was thinking though, that I needn't have some of those components (I
could quite easily be wrong), because I am just using this in a power
line, so I don't care about frequency or output ripple at all. Is
this correct? I found this site to calculate the values to use for
components to use with this chip (how lucky!) at
http://www.nomad.ee/micros/mc34063a/index.shtml . Now, if I leave the
fields blank for Vripple, and Fmin (12Vin, 40Vout, 175Iout), I get the
following:

Ct=NaN pF
Ipk=1254 mA
Rsc=0.239 Ohm
Lmin=NaN uH
Co=NaN uF
R=180 Ohm
R1=2k R2=62k (40V)

I don't know what NaN means...but is it implying that I can leave
those parts out? Also, it calculated 62k and 2k for resistor values,
which give exactly 40V (as opposed to the 39.2V of my configuration),
so it definitely beats me ;P.

Thank you very much,

Jacob Evans

--Oh, and so nobody gets mad at me, disclaimer:
I apologize if I have gotten any terms wrong..I just don't know them
that well, as I am relatively new to this kind of thing. If I said
anything wrong enough to make it unintelligible, let me know, and I'll
clearify. If I got anything wrong that you'd like to correct me on,
feel free to do so, it would be beneficial for me . Also, no, I am
not doing this for any kind of assignment, and want help...I'm just
doing it for a project for myself.

Beware.
The maximum switch collector to emitter voltage for this chip, is 40v. You
will be beyond the rating of this chip (by the drop in the output diode, and
half the maximum 'ripple'). As it stands, the circuit example, is not
suitable to generate 40v DC.
The calculator is giving 'NaN', because it cannot calculate values for these
parts, without some data to start with. Though you are not worried about
ripple, make a 'worst case' figure and apply this.

Best Wishes

 

[Face]

Beware.
The maximum switch collector to emitter voltage for this chip, is 40v. You
will be beyond the rating of this chip (by the drop in the output diode, and
half the maximum 'ripple'). As it stands, the circuit example, is not
suitable to generate 40v DC.
The calculator is giving 'NaN', because it cannot calculate values for these
parts, without some data to start with. Though you are not worried about
ripple, make a 'worst case' figure and apply this.

Best Wishes

Hi--
Thanks, for the warning--but to tell you the truth, I don't actually
know what this ripple is....what do you think that a good worst case
scenario would be? It isn't powering anything delicate that needs an
kind of precision.

My _guess_ is that Vripple is the ammount of fluctuation that can
occur in the Vout (in either direction?). So I tried a value of 500
mV, but it still cannot give me values, because it wants an Fmin
val...Since I really don't care at all, I arbitrarily put in a value
of 15kHz for Fmin (500mV Vripple, 12Vin, 40Vout, 175mA Iout), and this
is what I got:

Ct=1922 pF
Ipk=1254 mA
Rsc=0.239 Ohm
Lmin=422 uH
Co=151 uF
R=180 Ohm
R1=2k R2=62k (40V)

Would this be an adequate set of parts to use?

Thanks,
--Jacob

 

[Spehro Pefhany]

Hi,
For something I am working on, I need to have ~40VDC, where I normally
have 12VDC. I'm planning on getting this:
http://www.st.com/stonline/books/pdf/docs/5257.pdf . On page 6 of
that datasheet, it shows a circuit diagram which makes use of the
MC34063A, that outputs 28V with a 12Vin. I figured (using the formula
in the notes for calculations on page 9), that to get 40V, changing
the resistors to R1=2.7kohm & R2=82kohm should work. Now, should I
change anything else, or should that be fine?

I was thinking though, that I needn't have some of those components (I
could quite easily be wrong), because I am just using this in a power
line, so I don't care about frequency or output ripple at all. Is
this correct? I found this site to calculate the values to use for
components to use with this chip (how lucky!) at
http://www.nomad.ee/micros/mc34063a/index.shtml . Now, if I leave the
fields blank for Vripple, and Fmin (12Vin, 40Vout, 175Iout), I get the
following:

Ct=NaN pF
Ipk=1254 mA
Rsc=0.239 Ohm
Lmin=NaN uH
Co=NaN uF
R=180 Ohm
R1=2k R2=62k (40V)

I don't know what NaN means...but is it implying that I can leave
those parts out?

NaN = not a number. It means that there is an error in the
calculations. You will have to enter reasonable values for those
parameters to get values for those components. And no, you can't leave
them out.

Also, it calculated 62k and 2k for resistor values,
which give exactly 40V (as opposed to the 39.2V of my configuration),
so it definitely beats me ;P.

The absolute maximum value for the switch voltage is 40V, so maybe a
bit under 40V would be safer.

Best regards,
Spehro Pefhany

 

[Tim Wescott]

NaN = not a number. It means that there is an error in the
calculations. You will have to enter reasonable values for those
parameters to get values for those components. And no, you can't leave
them out.




The absolute maximum value for the switch voltage is 40V, so maybe a
bit under 40V would be safer.

Best regards,
Spehro Pefhany

If you're leaving out your ripple values there's a good chance that the
applet is defaulting them to zero. Since cap and coil values will be
proportional to the inverse of the ripple you'll get divide-by-zero
errors which often result in NaN results.

Surely you have _some_ ripple requirement? Is 40V ripple (i.e. your
output oscillates between 0 and 80V) too much? 20V? 2V?

 

[Face]

If you're leaving out your ripple values there's a good chance that the
applet is defaulting them to zero. Since cap and coil values will be
proportional to the inverse of the ripple you'll get divide-by-zero
errors which often result in NaN results.

Surely you have _some_ ripple requirement? Is 40V ripple (i.e. your
output oscillates between 0 and 80V) too much? 20V? 2V?

Okay..I understand..I'll probably use a value of a few volts, or maybe
less. But what about for Fmin? It IS just power, so should I just
specify any number, or maybe just one that gives my caps convenient
values?

Also, the Inductor at the top (I know very little about inductance, so
bear with me please ), is labeled as Lmin. Since it says "min" does
that mean that anything more will work as well (because it's sometimes
hard to find exact values)? Does it matter what type of inductor it
is, ie wrapped on a toroid?

Thanks again,
Jacob Evans