Maker Pro
Maker Pro

Low voltage, high current power supply

stube40

Feb 9, 2010
85
Joined
Feb 9, 2010
Messages
85
I need to design a power source capable of delivering high current (maybe 200 or 300A) at only 2V.

I thought I may be able to use TIs TPS40180EVM module - it has a lower current rating that I'm after, but you can stack them in a modular fashion.

However, I wondered if anyone had any better suggestions.

I'm more of a digital electronics guy, so designing from scratch is not a sensible move. Rather, something with a public domain reference design that I can just design into my own PCB would be ideal.
 

stube40

Feb 9, 2010
85
Joined
Feb 9, 2010
Messages
85
A VRLA battery.

Interesting. We have banks of AGM LAs here, so I could use them. The trouble is that they are 12V and I only need 2V. I suppose a single cell LA could get 2.3V or so, but I'd still need to be able to cap or regulate the current.

Am I thinking on the same wavelength to you or did you have something else in mind?
 

trobbins

Jun 15, 2010
83
Joined
Jun 15, 2010
Messages
83
You will need to provide a lot more information to allow a suitable response - especially on what power is available - how regulated - controlled range of V&I - what is load.
 

stube40

Feb 9, 2010
85
Joined
Feb 9, 2010
Messages
85
You will need to provide a lot more information to allow a suitable response - especially on what power is available - how regulated - controlled range of V&I - what is load.

The load is a coil made from a superconductor that has a virtually zero resistance and around 50mH of inductance.

From an available power source perspective, I have an array or 16x 12V 18Ah VRLA batteries available. Other than that, I have 240V mains (15A sockets) and 3-phase.

V range is 1V to 10V, I is 100A to 300A.
 

trobbins

Jun 15, 2010
83
Joined
Jun 15, 2010
Messages
83
Why do you need a V range from 1 to 10V when load has negligible resistance?

Do you have a tolerance on the set parameter of current, and any special requirements for the control of the current.

Is ripple an issue.
 

stube40

Feb 9, 2010
85
Joined
Feb 9, 2010
Messages
85
I am working on a new slant on the old homopolar turbine idea. I have a working prototype of the motor here. Using a heavy duty Lab DC PSU, when I apply 2.0V to the motor it draws 100A at rotates at approx 2000rpm.

At 3.5V it draws 180A and rotates at around 3000rpm.

I need to hook up a torque transducer to the output to equate this to power (still trying to beg/borrow/steal one at the moment).

The main thing for me at the moment is to replace the Lab DC PSU with my own design.

I'm not really sure about ripple to be honest, althoughI have use a scope to plot the current and voltage response when the motor is running. I could post these if it helped?
 

trobbins

Jun 15, 2010
83
Joined
Jun 15, 2010
Messages
83
Umm, why is the lab PSU deficient?

So your powering a brushed DC motor?
 

stube40

Feb 9, 2010
85
Joined
Feb 9, 2010
Messages
85
Umm, why is the lab PSU deficient?

So your powering a brushed DC motor?

There's nothing wrong with the lab PSU - however, there's some good technical reasons why I need to be able to make my own version of it (eg control, integration into existing system, cost, power efficiency).

Yes, it's a brushed DC motor
 

trobbins

Jun 15, 2010
83
Joined
Jun 15, 2010
Messages
83
You initially pointed to a sync buck evaluation module. I strongly suggest that unless you want to take up another vocation as a switchmode designer, that you take one step back and look for a colleague/consultant to do the design for you as a full-on project (ie. if you wanted to spend the effort-cost to prepare a new product), or look at how to patch together commercial products.to get you something suitable.

Maybe you can justify the effort to prepare a suitable PSU, but just go in with your eyes open as to the effort and experience and time that will be needed.

You may find some suitable switchmode electro plating rectifiers that have voltage/current ratings that suit your application. Jewellers use that type of gear.

Ciao, Tim
 

stube40

Feb 9, 2010
85
Joined
Feb 9, 2010
Messages
85
Thanks Tim,

I have an experienced electronics engineer I can use one we get to the nitty gritty - he's good at 'filling in the blanks' once someone has come up with the overall solution. However, I wouldn't want to give him a free reign with proposing the type of solution since I'll end up with something 'star trek'.

Hence the reason I'm getting involved at this level (I'm a mildly experienced digital electronics guy come embedded software engineer come supervisor).

Cheers,
Stuart.
 

trobbins

Jun 15, 2010
83
Joined
Jun 15, 2010
Messages
83
You still have a significant effort, and a learning curve for your elec eng if he hasn't done many switchmodes before.

To minimise the effort, you can just do dc/dc at low voltage and use a commercial power supply to provide (say) the 12V grunt, and a battery to buffer any interaction. Then you get to something like the sync buck option you identified. It is as good an option as any - aim for the most support possible, such as the evm to get the basic performance and parts selection covered. You still need to review the evm to see whether it covers your application.
 

stube40

Feb 9, 2010
85
Joined
Feb 9, 2010
Messages
85
Thanks Tim,

Hopefully this is the last time I bother you on this topic..........

OK, let me just check I've got this right. The chain is

1) 240 AC to 12V DC by commercial PSU
2) Battery to buffer any interaction
3) 12V to 2V by DC-DC

Is that right?

I'm intrigued by your use of a battery as a kind of filter - how does that work and what is the connectivity to it? If I put it in series then I'm simply adding another +12V to the 12V output of the PSU? Also, were you referring to using a LA battery?

Thanks very much,
Stuart.
 

trobbins

Jun 15, 2010
83
Joined
Jun 15, 2010
Messages
83
Your chain is what I had in mind.

You may get transient and feedback loop problems if you connect the output of one switchmode, to the input of another switchmode - but you may not! If not then no need for a battery, or a largeish capacitor. The battery or capacitor would sit at the common 12V at the input/output of the tow PSU's. Probably best to sequence the PSUs, with the Mains-to-12V turned on first, and off last. It may be useful to include a series diode in the output of the mains-to-dc PSU (if it doesn't include one), to provide fault isolation.
 

stube40

Feb 9, 2010
85
Joined
Feb 9, 2010
Messages
85
Hello Tim,

An internet search has drawn a blank regarding a DC-to-DC converter that can deliver the high currents (100s of amps) whilst stepping down from 12V to 2V.

Did you make that suggestion knowing that there were products that could do it, or is it possible that I've stumbled across a flaw in the plan?

Thanks,
Stuart.
 

trobbins

Jun 15, 2010
83
Joined
Jun 15, 2010
Messages
83
Stuart, what was wrong with the evaluation kit you proffered initially - as it seemed to cover what you needed pretty closely. You may be able to increase the current rating of a module, and I doubt uping the voltage to 2V will cause any concerns. Like all things, you would need to go through an in-depth review to appreciate whether you can get suitable parts to emulate the design yourself, and what part choices limit your output current rating. There may be higher level specs in that EVM than you need, which may assist.

If the controller is not a practical choice for you, then the other option is to revert to a using a more available sync buck controller and use the EVM as a guide to design a single stage 'beast'. I'm sure it could be done, but you soon appreciate that part selection and availability and constraints start hitting you at each turn - which can be an incentive to plow through, but perhaps only if you have the experience to know what your battling.
 

stube40

Feb 9, 2010
85
Joined
Feb 9, 2010
Messages
85
Thanks Tim,

Yes, the original equipment I mentioned would now seem a reasonable choice.

I will investigate this further to see if it can be modified sutiably. Thanks for all your help.

Stuart.
 
Top