Question About DC-DC Step Up Converter

Wouldn't it be simpler to get a few more batteries and wire them in series to desired voltage?

Unfortunately that is not an option as space is very limited.

Interesting lol... Hadn't thought that one through. Think these things could actually deliver momentary loads of say 7.5 amps for 100ms or so in series?

The LiPo probably would....but not very many times I think.....You can always go to a larger capacity battery pack, but power density vs space available, is not going to change no matter what type of electronics you have....In other words you either have the space required, or you don't....

Put a *large* capacitor across the batteries. 7.5 amps for 100 ms is 0.75 coulombs of charge. The voltage will change from its quiescent value to a lower value the battery can sustain under that load. If you know how much your voltage can swing over that interval (and still remain within spec from the controller's standpoint) you can estimate the capacitance you'll need.

This brings up an interesting idea....what about a capacitor bank?

What charges the capacitors? Their only advantage is to accommodate high peak current, as they can discharge at a much higher rate than a battery. Capacitors have a very low energy density compared to batteries. Joules vs kilojoules in a much smaller form factor.

There are so many similar but different graphs on the energy and power density of batteries and capacitors that I couldn't choose which to post. I won't beg for more hidden data but I suspect the motor drivers desire a relatively narrow voltage range to work. This will leave most of the stored energy in a capacitor.

Put more batteries in series.

[Now I'm out of here.]

<unsubscribe>

Why would you need to beg for more data? I'm a student, and I have a job to attend to during the day. I'm trying to be as forthcoming as possible...

Anyway, the motor controller will operate the motor anywhere between 24vDC - 80vDC, but as the drive voltage decreases the maximum torque and maximum speed the motor can achieve decrease as well. I'm trying to keep the speed and torque which is why I want to run it at 72vDC. I understand this is hard to do, but I've been trying for some time and was just looking for some suggestions.

Here is a link to the controller spec: http://www.linmot.com/fileadmin/doc/InstallationGuides/Installation_B1100_e_recent.pdf

"Beg" for more data? No one here is begging you for data or accusing you of not being forthcoming. At least I'm not.

Let's turn this around, shall we? Someone approaches you with this question, giving the same info you've supplied thus far and you would have no other questions? Surely not.

Your application plays a huge part. If your servomotor is to spin a constant load at a constant speed is one thing, but a servomotor driving the arm of a manufacturing/assembly robot is a very different matter altogether and will *greatly* impact the type and capability of power supply you use. It may be the same motor, but how you plan to use it figures significantly in the kind of answers you'll get.

550watts is a pretty hefty supply to design. Especially boosting 18v to 72v.

I think Solareagle's suggestion of using more batteries is the best/easiest/quickest solution (and outside or charging, it's fool proof).

You never indicate the amp-hour rating of your battery, or what the DC motor does.

If your stuck with this solution, then look at Linear Technology SMPS web site.

The battery has a 5 amp-hour rating and the motor isn't always on. Thanks for your suggestion.

These guys have a couple of options.

http://www.linear.com/products/step-up_%28boost%29_regulators

These are interesting, typical applications in the data sheets don't show anything but how to step down voltage to about 5v dc or so...

Here's something you may have overlooked, your switching supply is driving a servo motor. The back EMF from this motor may not have a safe reverse current path to flow when you stop or slow down this motor. Remember also that the time constant for discharging a charged inductor is τ=L/R. A very high reverse current path will discharge quickly but by making a very large voltage. This voltage can easily exceed the breakdown voltage of many semiconductors with catastrophic effect.

BEMF is quenched by the servo controller and doesn't effect the power supply.

Lol, I'm not trying to hide anything. The servo controller is off-the-shelf and I've talked to their tech guys who say it can be done via battery power they just don't offer a solution...

What motor and controller are you using?

Servomotors can draw very high peak currents, depending on the application. Your power supply will need to handle this with margin to spare. Designing a suitable power supply is *not* trivial. Have you looked for off-the-shelf solutions? Have you considered powering an AC-sourced DC power supply from an inverter? Is your application portable? I'm guessing that it is either portable or will be used in a remote location where mains power is unavailable. Those MOSFETs? How did they burn out, and under what conditions? I'm presuming these were in some sort of power supply? Did they burn out under load? Did they work before you connected the servo controller? When you started the motor under no-load condtions? When you reversed under full-load?

We're not there. You are. Tell us what you know.

Here's a KISS thought...

An off-the-shelf 1KW automotive 12VDC->120VAC inverter might work OK with 18V - and not be too pricey. Plug a 120VAC->72VDC power supply into that and you're in business!

Me, I'd run my AC from the inverter through a VARIAC so I could tweak the DC down to the minimum required to do the job - and yes, use BFC's for the start-up surge current.