24V Vehicle PSU / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

24V Vehicle PSU

1 2 next

Move Thread

LAN_403

Dave Marshall

Mon Feb 23 2009, 08:01PM

Registered Member #16 Joined: Thu Feb 02 2006, 02:22PM
Location: New Wilmington, PA
Posts: 554

I'm in need of a 24V, 10A power supply that will be powered by the 12V DC bus in my truck. I know there are a number of ways to go about designing such a system, but I have a significant hurdle to overcome.

I have an HF radio installed in the truck that is absolutely essential. I operate on frequencies from 3.5-30MHz, and RFI is a major issue.

This eliminates most SMPS topologies from consideration, as they are a nightmare of harmonics.

I have a large bank of 12v batteries in the bed of the truck to allow the radios to operate without depleting the truck's battery when the engine isn't running. These batteries can be utilized for the 24v rail if need be.

The purpose of the 24V supply would be to allow the use of RF friendly linear MOSFET regulation to voltages suitable for two laptops (~18V) and a couple other computer peripherals that want more than 12V.

I had considered simply creating a very clean sine wave <500Hz and driving 12V into 500W 120V-240V transformer, but I suspect the result would be a good bit less than 24V, necessitating a custom rolled transformer. Regardless of how clean the drive signal is, there would likely still be harmonics, also necessitating very large filters.

If some of you solid state gurus have any ideas, I'm certainly all ears.

Dave

tesla500

Mon Feb 23 2009, 08:18PM

Registered Member #347 Joined: Sat Mar 25 2006, 08:26AM
Location: Vancouver, Canada
Posts: 106

With proper EMI filters, pretty much any topology should work. I'd suggest a simple boost converter, built keeping EMI in mind. That means building the supply in a completely shielded case, with good common and differential mode chokes at both the input and the output, and following good EMI practices. Sheilded IO cables would also likely help.

What about your truck's ignition system, or computer? Those produce EMI too, and they're presumably filtered well enough that your radio still works. You also mention laptops/computer peripherals, those have high frequency buck converters in them to generate various supplies too, if they can run with your radio, so can a good boost converter.

David

Dave Marshall

Tue Feb 24 2009, 04:34AM

Registered Member #16 Joined: Thu Feb 02 2006, 02:22PM
Location: New Wilmington, PA
Posts: 554

The computers have been thoroughly vetted, and produce minimal problems themselves, especially when the lids are closed and the screens are off (one of them will always be in this state). The other luckily is just RF friendly in any state (except when the factory PSU is used).

The vehicle is an '08 Tundra, and is considered one of the best vehicles available in terms of mobile radio installation and EMI. I've already shielded ignition wires, along with EMI filters on both the 12v bus into the cab and the ignition itself. It is considerably quieter than my home station from an RFI perspective.

The big problem is that a shielded box is probably not going to do a whole lot of good, because I have no proper ground.

The radios and antennas are grounded to the same truck body/frame that said shielding would connect to. This will lead to a rather efficient noise distribution system. I'll be shielding and grounding as best I can, but really the ideal solution is to keep noise to a minimum to start with, and work from there.

Dave

GeordieBoy

Tue Feb 24 2009, 01:37PM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

Hi Dave,

As I see it there are two options. You could either series connect a stack of batteries to get the 24V you require or use some fancy power electronics to convert 12V up to 24V.

The series connected batteries is a no brainer so no explanation necessary.

There are a number of ways you can use power electronics to step up 12V to 24V. Unforunately they all involved chopping up the power at high frequency and smoothing it again. However some choices of converter are much more wise than others.

The boost converter is quite a good choice in terms of EMI because it's input current is continuous (i.e. not chopped on and off by the switch) This means that EMI conducted back out of the input side to the vehicle's 12V battery will be very low. The downside is that the output current is discontinuous so more conducted EMI is likely on the output port of the boost converter and filtering is likely to be needed here as this feeds your sensitive RF equipment.

Don't even consider the flyback converter as this has discontinuous input current and discontinuous output current. Therefore it spits conducted EMI out of both ends, and in large amounts if not filtered. The high di/dt in the flyback converter and gapped transformer core are also a recipe for very strong radiated magnetic fields.

A buck derived converter may be a good choice as its output current is continuous at full-load when designed correctly. The downside is that the input current is discontinuous but this can usually be easily filtered with a combined common-mode / differential mode filter and a big electrolytic cap at the input to the converter. Since you want 24V out for 12V in, the best choice of converter would be a push-pull transformer coupled converter where both primary side MOSFETs are grounded and the centre tap of the transformer is connected to the +12V bus. The turns ratio can be adjusted to get +24V output after rectification and smoothing. Closed loop regulation can then be applied depending on how accurately this 24V needs to be held.

I wouldn't consider that Switched Mode Power Supplies aren't an option. They can be made to be very quiet in terms of EMI it just takes a bit of effort and a little experience in this area too.

I know Steve Conner runs an HF rig from a Switched Mode Power Supply that he modified for the purpose. I beleive it is a good quality supply from someone like Astec or Lambda though, so you would expect it to be fairly quiet in EMI terms.

My recommendations would be to go with the boost converter if isolation between input and output isn't required, or to go with the push-pull converter if isolation is required. Put the whole thing in a sealed metal box, and have seperate partitions for the input and output filters. Use a relatively low switching frequency like 40kHz, and allow compromises on size and efficiency in order to slow down the switching transitions and reduce EMI. Noise should not be conducted out of the shielded box if the input and output filters are implemeted correctly. The idea is to make a faraday cage and filter the hell out of any wires that breach the walls. Finally snubbers across the switches and diodes should tame any HF ringing before it has chance to couple to the wires leaving the box.

-Richie,

Steve Conner

Tue Feb 24 2009, 05:16PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

I'd suggest you not bother with linear post-regulation and just learn to build quiet DC-DC converters. You already have laptops in your system, you're presumably happy with their RF performance, and they're packed full of DC-DC converters of all kinds, so it can be done. And the skill will come in handy in future.

I'd start by buying a regular cigarette lighter-to-laptop adaptor suitable for your laptop, and transplanting the guts into a diecast metal box. Use feedthrough capacitors for the DC + input and output terminals (the box itself comprises the - input and output) This forms a complete Faraday cage in the way that Richie describes.

If this doesn't give enough suppression, then you build the feedthrough caps into pi filters, by adding an inductor and capacitor on the inside of the box at each terminal. You can actually buy complete canned pi filters for feedthrough mounting, but they can be a bit expensive.

Yes, I did convert a SMPS to run my HF rig. It was a good industrial grade one to begin with (an Astec LPS250) and I built a custom LC filter for the DC output, enclosed in a copper box, that bolted onto the front of it. It still emits some noticeable QRM on 160m and 80m, though. I plan to try adding an extra filter to the mains end of it next.

Bored Chemist

Tue Feb 24 2009, 05:58PM

Registered Member #193 Joined: Fri Feb 17 2006, 07:04AM
Location: sheffield
Posts: 1022

Could you charge a couple of 12 volt batteries then connect them in series to give a 12 v power supply? A couple of car batteries is about as smooth and low source impedance a supply as you can get. You might need a couple of switches or relays but it's a whole lot easier to design than a low noise SMPSU.

Dave Marshall

Tue Feb 24 2009, 08:24PM

Registered Member #16 Joined: Thu Feb 02 2006, 02:22PM
Location: New Wilmington, PA
Posts: 554

I had thought about simply charging the batteries in parallel and discharging in series, but there doesn't seem to be any way to do this continuously. It would require charging a bank and then discharging them, and given the capacity of the batteries, they'd become exhausted rather quickly.

I currently have 60AH available in the form of 4 15AH SLAs. I have two 105AH marine batteries as well that could go in the back of the truck instead, but they weigh nearly twice as much as the SLAs, and combined I'd be adding something close to 300lbs in the back of the truck.

Not a huge deal, but definitely a pain in the butt. really what I need is a marx generator type doubler, it just doesn't seem to be possible.

Dave

Sulaiman

Tue Feb 24 2009, 09:35PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

Could you consider the passenger compartment 'noisy' due to
your computers and any other gear, and screen the rig from the vehicle?

As far as possible put metal around the rig.
This would also screen the compartment from your rig !

GeordieBoy

Wed Feb 25 2009, 12:04AM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

Another option might be a switch-mode charger to charge a 24V battery pack from 12V when NOT using the radio. This can then be switched off to reduce EMI while the radio is being used?

Dave Marshall

Wed Feb 25 2009, 03:11PM

Registered Member #16 Joined: Thu Feb 02 2006, 02:22PM
Location: New Wilmington, PA
Posts: 554

That may turn out to be my only real option. At 60AH, I should exhaust the battery bank in about 5-6 hours if both laptops were drawing maximum power (only one likely will be doing that, the other won't even have the screen on). If I split the bank into two 30AH sets, I should be able to recharge one while using the other, allowing me to draw that out atleast a bit, perhaps giving me up to 8 hours before both banks would be completely exhausted.

This is for the vehicle I use for storm chasing (see my latest post in the Cool Picture thread). One laptop is for mobile internet and video editing/uploading, the other is for recording audio from the 4 radios onboard the truck.

I can use a standard AC inverter when the radios aren't in use, so before and after the chase. Typically the really active phase of the chase where HF reception is a must only lasts 3-4 hours, so that seems like a reasonable option.

Dave

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob