[PSUBS-MAILIST] power converter

Wed Apr 27 09:39:44 EDT 2016

Hank - these motors of yours are DC motors, correct?  If so, the only way to control them is by varying the applied DC voltage, either directly or by approximating it with pulse width modulation (a DC motor controller) which outputs the full rated voltage (120 V?) for some percentage of time (the duty cycle), at a period switched at a relatively high frequency.

Either way, you need to step up from your 48 VDC bus voltage. There are three ways to do that. 

1). You can go to more batteries in series for a higher DC bus voltage. I like this solution because you get away with smaller wires and have less voltage drop over distance, but it does mean that you need to use regulated power supplies for house loads instead of making direct battery connections to 48 VDC appliances. Also, DC voltage is dangerous stuff, so make sure the bus lines are protected from contact, fused or with appropriate breakers etc.

2) You can use a DC-DC converter of an appropriate power rating for your motors. Not sure what it would cost, but this is the simplest and cleanest solution.

A converter is essentially a VFD in inverse. To step up any voltage, that voltage must be AC (transformers only work with varying electric / magnetic fields). A transformer operating at typical AC mains frequencies is large and inefficient. As the AC frequency goes up, the size and heat dissipated in components goes down. I'm simplifying a bit, but the DC-DC converter essentially uses some complicated switch mode electronics to invert your 48 VDC to an AC voltage at a very high frequency (kHZ or MHz), then steps that AC voltage up in what is effectively a transformer circuit (to e.g. 170 VAC), and finally rectifies the stepped-up AC voltage back to DC (e.g. 120 VDC).

3) Use a M/G set. 48 VDC motor directly connected to an AC generator and rectifier to produce DC at the higher voltage. Old school and inefficient, but it works.

At this point, I might just look at the cost of replacing the motors you have with 48 VDC equivalents.

FYI, in my own design, I am using a high voltage DC bus (647 VDC), but that voltage only exists in the battery compartment which is an isolated space flushed with sulfur hexafluoride in order to eliminate fire risk and prevent arcing. The battery feeds the DC bus of a VFD, bypassing the rectifier stage which would otherwise be fed by an AC generator or shore power for charging. The VFD outputs variable frequency AC at 440 V, feeding an asynchronous induction motor connected to a hydraulic pump, which outputs to a common pressure line in parallel with a number of accumulators for energy storage. All of the vessel thrusters and actuators are hydraulic, commanded by servo or proportional valves as appropriate. Only 24 VDC is present in the manned compartments. The hydraulic system is inefficient compared to pure electric, but capable of delivering high power in small form factors. I get around some of the efficiency problem by relying on the accumulators so that the pump isn't running
all the time.

Sean

On April 27, 2016 6:24:26 AM MDT, hank pronk via Personal_Submersibles <personal_submersibles at psubs.org> wrote:
>Hi Kieth,If I use my 3,000W  inverter do I not need a  rectifier?
> could I still reverse them?  Thank youHank 
>
>On Tuesday, April 26, 2016 9:48 PM, k6fee via Personal_Submersibles
><personal_submersibles at psubs.org> wrote:
> 
>
>Most inverters these days are 90 - 98% efficient at converting DC to
>AC. You would only need a 100,000 mfd cap to handle smoothing the DC
>ripple to acceptable levels to 50 amps and they rarely go south
>(fingers crossed) in my experience.
>With that said having 120vdc direct from a battery bank is the
>preferred  option.
>Keith T.
>
>
>Sent from my Verizon, Samsung Galaxy smartphone
>-------- Original message --------From: Stephen Fordyce via
>Personal_Submersibles <personal_submersibles at psubs.org> Date: 4/26/16
>7:45 PM (GMT-08:00) To: Personal Submersibles General Discussion
><personal_submersibles at psubs.org> Subject: Re: [PSUBS-MAILIST] power
>converter 
>Hi guys,Getting it to AC should be more straightforward in terms of
>being able to get a standard product, but still needs a pretty
>big/expensive inverter.  The DC rectifier could work but again for the
>currents involved the capacitors required for a smooth output would
>have to be huge, and it's another significant exercise/stuff to go
>wrong.  Plus it's probably all quite inefficient (I wouldn't be
>surprised if you lost 20% of the electrical energy just through the
>conversion).
>Cheers,Steve
>
>
>On Wed, Apr 27, 2016 at 12:18 PM, k6fee via Personal_Submersibles
><personal_submersibles at psubs.org> wrote:
>
>Another option is to use an off the shelf 48v to 120v ac inverter then
>use a bridge rectifier to take it back to DC. To get away from a
>cooling fan the heat sink could be mounted to the hull to dissipate the
>heat and make it quiet.
>Keith T.
>
>
>Sent from my Verizon, Samsung Galaxy smartphone
>-------- Original message --------From: Stephen Fordyce via
>Personal_Submersibles <personal_submersibles at psubs.org> Date: 4/26/16
>6:30 PM (GMT-08:00) To: Personal Submersibles General Discussion
><personal_submersibles at psubs.org> Subject: Re: [PSUBS-MAILIST] power
>converter 
>Hi Hank,I've dabbled a little bit in (small!) DC converters and my
>guess is that a converter that will do the current you're after will
>be expensive/large/complex and probably generate a significant amount
>of heat.  Also seems like an obscure set of operating conditions that
>will be hard to find off the shelf.  
>You've probably thought about it, but splitting the battery bank
>and rewiring part of it to supply 120VDC is probably the easiest
>low-tech solution if you have enough cells to put in series and achieve
>the higher voltage.  You would get some nice redundancy there too if
>you have 2 battery banks controlling different thruster sets.
>As an aside, I suspect 48VDC is pretty lethal, but 120VDC even more so.
>I just saw Alan's email come through - running it off lower voltage is
>definitely an option, but depends on the sort of motor.  Ie. a brushed
>motor speed is proportional to the voltage applied (so running at 48VDC
>would give you less than half speed) - no idea what the Perry ones
>would be, but being older they may well be brushed.  Brushless motor
>speed is proportional to frequency applied by the controller.  In both
>cases torque is proportional to current which is proportional to heat
>generated.  So the greater the current in your wires, the more heat
>needs to be dissipated, and if it's not then something will burn out -
>probably the thin layer of insulation on the motor windings.  The
>trouble is dissipating the heat from the inner windings (it has to pass
>through the outer windings, which are producing their own heat) - but
>like Alan said, if you've got it oil compensated that will help a lot
>with heat dissipation.  The thermal conductivity of oil is ~6x that of
>air (and water ~24x air), although the heat transfer is not quite as
>simple as that.
>I'm also pretty sure that the motor will have an inherent
>resistance that will be designed into it appropriate to the nominal
>voltage and current, and related to the size of the wires (ie. as Hank
>pointed out, thinner wires for high voltage as less current is
>required).  So if you apply less voltage to the same resistance, it
>will draw less current - ie. I don't think you will be able to get
>extra current into the motor at lower voltage.  Not quite that simple,
>but you get the idea.
>(that was mostly dredged up from memories of a university subject I did
>on electric motors a long time ago, feel free to correct me!)
>
>Cheers,Steve
>On Wed, Apr 27, 2016 at 10:18 AM, hank pronk via Personal_Submersibles
><personal_submersibles at psubs.org> wrote:
>
>Thanks" Kieth,I started to that but I have no idea if one system is
>better than another.Hank 
>
>On Tuesday, April 26, 2016 6:11 PM, k6fee via Personal_Submersibles
><personal_submersibles at psubs.org> wrote:
> 
>
> Hank,
>You want a dc-dc converter, just Google it.
>Keith T.
>
>
>Sent from my Verizon, Samsung Galaxy smartphone
>-------- Original message --------From: hank pronk via
>Personal_Submersibles <personal_submersibles at psubs.org> Date: 4/26/16
>4:53 PM (GMT-08:00) To: Personal Submersibles General Discussion
><personal_submersibles at psubs.org> Subject: [PSUBS-MAILIST] power
>converter 
>HELP,,,Turns  out my Perry thrusters are not 36V but 120 Vdc.  I can
>only manage 48V dc from Gamma's battery bank.  Can I step the voltage
>up with something.Hank
>_______________________________________________
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
>
>
>   
>_______________________________________________
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
>
>
>
>
>_______________________________________________
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
>
>
>
>
>_______________________________________________
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
>
>
>
>
>------------------------------------------------------------------------
>
>_______________________________________________
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles

-- 
Sent from Kaiten Mail. Please excuse my brevity.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.whoweb.com/pipermail/personal_submersibles/attachments/20160427/d25b0c1d/attachment.html>