High current boost converter?

[electronwaster]

Hi,

I'm completely new to the blog - and fairly new to electronics, but this place seems like a great place to ask questions and I really enjoy watching the video blog.

One of my current projects is to build a laptop power supply for the car. I know they are available to buy, but I'd like to try and build one for the experience of it. I watched the tutorial on the MC34063, and this seems like the best chip to use, especially as I have a few in my growing parts bin.

I have looked at the datasheet a while ago, and I believe they show using a pass transistor when using a high current output, but I couldn't find any guide as to which transistor to use, or how to modify the list of formulas that I went through (and that Dave shows in the vid).

This information could do double-duty for me, as I'd also like to build a multi-output high current USB charger. So - for example - 3 or 4 5v outputs at 1A or greater each. I realise this may need some heat calculations and large heatsinks added!

If anyone can shed any light or show a working circuit for a high current MC34063 circuit or equivalent, I'd be very grateful!

Thanks,
electronwaster

[IanB]

You might be taking on something a bit complicated for a first project...

However, before thinking about a design you should define your requirements. For example, my laptop power supply provides 19.5 V at up to 4.6 A. If you wanted to do that from a car, you would need a boost converter consuming about 10 A from the car's 12-14 V electrical system. That is going to be a very substantial circuit with big components and a fair chance of things going "poof!" if you make a mistake.

It is possible perhaps that this may not be necessary. It may be that a laptop expecting 19.5 V will still work at reduced speed if provided with 12 V. You could try it and see. No voltage converter may be necessary.

One thing to watch out for is noise and spikes on the car electrical system. A big part of your design is going to be filtering and surge suppression to prevent noise and excess voltages causing damage or malfunctions.

[amspire]

Ian's advice is very sound.

Boost converters only limit the output voltage if they are regulating properly. If you make a mistake, you can easily exceed the laptop's maximum input voltage and blow it. Also, it is very likely that the boost converter will generate electrical noise that will interfere with the car stereo.

Laptop power supplies for cars are readily available at good prices, and they have cases custom designed to dissipate the heat from the converter.

I would probably look for a less challenging first project.

Richard

[electronwaster]

This is not my absolute first project, I have made a variable speed extractor fan with an LM317, modified an antec PSU for use as a bench power supply, and obtained an old kikusui scope, a copy of art of electronics and all the futurlec value bags and some other goodies. I have been trying to read and trying to understand for the last 4-8 weeks - but like the recent poster here, I feel like I have hit a dead end and my eyes glaze over when it comes to (in particular) maths, impedance, complicated bits in general. Some things just seem to leave me scratching my head, so I'm attacking from a different angle and trying to get some projects made. I think I have the hang of resistors, ohms law, high and low pass filters, and I know how and why to limit current into an LED, etc. I have a basic understanding of transistor amplifiers, but couldn't design one. That's more or less my experience level, anyway.

My laptop is really a netbook. The power supply is rated to provide 19V at 2.15A - I realise this may not strictly be what the laptop needs in terms of voltage or current, but I'd really like to treat those numbers as the specifications for my project and not "cheat", and so, build something that meets those numbers. 19V*2.15A is about 41w, and these switching supplies I believe are around 80-90% efficient, so I'll be drawing 45-50W,  which at 12V is about 4 or so amps. I think the cigarette lighter socket in my car and van are each rated to 10A (fused at 10A anyway), so I should be fine there. I have no problem running a heavier cable to the cabin and fusing it myself if necessary though.

Really I'm looking for a set of formulas like the datasheet ones Dave goes through in the video, but ones that are designed around the use of an exterior transistor, and then a specific transistor to use! Otherwise, I might have to look at another IC, which would be a shame, as I bought a small tube of these things, thinking they'd be good for general purpose power supplies.

Thanks,
electronwaster

[electronwaster]

Boost converters only limit the output voltage if they are regulating properly. If you make a mistake, you can easily exceed the laptop's maximum input voltage and blow it. Also, it is very likely that the boost converter will generate electrical noise that will interfere with the car stereo.

I thought boost converters were pretty reliable and already in use for things like laptop power supplies etc. - or when you say regulating properly, do you mean inside a properly designed circuit?

Laptop power supplies for cars are readily available at good prices, and they have cases custom designed to dissipate the heat from the converter.

I was hoping to sort out both the electrical noise and heatsinking problem by mounting the circuit inside a die-cast project box - do you think that would be sufficient for the noise? I realise either some calculations or experiments have to be done to decide whether it will work as a heatsink.

Thanks,
electronwaster

[Psi]

To give you an idea, this is a powersupply i built many years ago to take 10-14VDC and output 12VDC at up to 10A.
(well... 10A on a good day and with 14V input. Definitely 8 amps over the input range)

The chip is a TL494 and the 9volt is in their for size comparison.

Note: It only works at around 40khz, so could be much smaller if it was redesigned today with a higher switching frequency and newer low esr caps.
I don't recommend you copy my circuit (there maybe errors i forgot to update in the diagram)
but the TL494 is still a common IC for high current push/pull powersupplies with lots of info available.

[amspire]

Boost converters only limit the output voltage if they are regulating properly. If you make a mistake, you can easily exceed the laptop's maximum input voltage and blow it. Also, it is very likely that the boost converter will generate electrical noise that will interfere with the car stereo.

I thought boost converters were pretty reliable and already in use for things like laptop power supplies etc. - or when you say regulating properly, do you mean inside a properly designed circuit?

Yes. If there is a mistake in your circuit, or the regulation can get unstable, then the boost converter can generate very high voltages.

It is just a warning of what could happen. If you build the circuit,  I would start off with some nice big power resistors as a dummy load until you have really tested the circuit out.

Laptop power supplies for cars are readily available at good prices, and they have cases custom designed to dissipate the heat from the converter.

I was hoping to sort out both the electrical noise and heatsinking problem by mounting the circuit inside a die-cast project box - do you think that would be sufficient for the noise? I realise either some calculations or experiments have to be done to decide whether it will work as a heatsink.

Thanks,
electronwaster

The majority of the noise will be in the input and output power cables.  Minimizing noise is a matter of minimizing spikes and ringing in the switching circuit, getting the layout correct for lowest noise, choosing the right components, and getting the input and output filtering circuits right.

A perfect circuit and the right components with a bad board layout could make a very noisy power converter.

Richard

[Psi]

Yeah, i agree about the layout being important.
Be sure to use a star ground type layout.

[IanB]

To give you an idea, this is a powersupply i built many years ago to take 10-14VDC and output 12VDC at up to 10A.

In the picture I see two inductors (lower left and upper left), but on the circuit diagram I see only L1 (I'm guessing that's the the gold one lower left?). Is the other one with the red wire a noise filter to suppress interference?

[Psi]

To give you an idea, this is a powersupply i built many years ago to take 10-14VDC and output 12VDC at up to 10A.

In the picture I see two inductors (lower left and upper left), but on the circuit diagram I see only L1 (I'm guessing that's the the gold one lower left?). Is the other one with the red wire a noise filter to suppress interference?

Correct, the red one is a noise filter that i added later.

[electronwaster]

Thanks for the comments, and the schematic for the old-style low frequency switcher.

So - basically it's clear to me that I will have to look into board layout and noise elimination (or reduction) a little more than I already have - putting it inside a metal box won't cut it. I guess when I've found the right circuit, I will submit a PCB layout for criticism here on the forum and see what the experts think!

I am still at a loss to which components to try and use with the MC34063, I would have thought that a google search would turn up more than it has. Maybe it's a situation where intuition or experience is relied upon rather than a go-to formula or rule-of-thumb.

I've been looking at other chips and the MAX668 looks like it might do what I want out of the box as far as voltage and current goes. It's a more expensive chip, but that's not too important for a one-off educational project.

Thanks,
electronwaster

[mobbarley]

national semi's web bench is a good place to see what a power converter like this might require.