Voltage spike protection

[ccnw]

It's ounce of prevention time. Having spent money on some new electronics for the boat (depth sounder/chart plotter combo, VHF radio) I'm concerned about hooking them up directly to the starting/deep cycle battery that's kept charged by my old (1985) outboard. It seems to be voltage regulated, but I'm worried about possible spikes and transients, especially for the notoriously sensitive GPS module. I'm wondering if a buck/boost DC to DC convertor like this (http://www.ebay.com/itm/181516035122?_trksid=p2060353.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT) will prevent spiky stuff from coming through. The recharging current comes off a coil under the flywheel on the top of the outboard, so I don't know if this type of source generates the kind of peaks I read about occurring in automobile alternators. Alternators I'm sure produce way more current than the outboard setup. Anyway, if anyone has insight I would appreciate it. (Yes, there are marine "power conditioners" but they are pretty spendy and violate the DIY credo..." Thanks!

[tautech]

Welcome to the forum.

You could just use an LC network, maybe with some additional fast capacitance, Ceramic or Tantalum as much as many dislike them.
Consider the battery as a good suppressor of spikes, they are absorbed well and providing your equipment supply comes from a seperate connection, not the charge cable all should be well.

But considering the equipment is made for marine installation, why don't you think manufacturers haven't allowed for spikes etc in their design?
What is recommended in their installation guides?

[ccnw]

    It seems to be a pretty widespread problem of spikes killing navionics, especially GPS modules. They're often wired to be on continuously. It sounds like the problem is commonly caused by spikes created by the starting motor coil creating a kickback of voltage from the inductance of the motor windings. I don't know why marine electronics would be more sensitive than automotive electronics. Though I'm sure regulations demand that automotive stuff be pretty bombproof. I wonder if the robustness of marine electronics hasn't kept up with the increasing sophistication and sensitivity, specifically with the now widely available consumer grade navionics. I'm sure the commercial stuff is way tougher- you're not going to be out in the Bering Strait with equipment that will crap out on you (I live in Seattle and you see a lot of heavy duty stuff down at Fisherman's Terminal.)
     In the consumer grade stuff damage caused by power issues is expressly not covered by warranty....

[Ian.M]

The problem with automotive alternators is that the regulation is done in the alternator by controlling the field current.  The field coils have a considerable self-inductance and as a result the alternator cannot respond quickly to a step change in the load current.  All is well as long as the alternator is properly connected to a large enough battery that's in good condition, but if the battery becomes disconnected (or develops a high resistance connection) while running, turning off the headlights or other high current load will cause an older alternator to generate a load dump pulse of the order of 100V and 1/2 second duration which will fry any unprotected electronics.  Newer vehicles with ECUs and other critical electronics usually have effective load dump pulse clamping either in the alternator or as a separate module.

Your outboard alternator almost certainly does not have a field coil, but uses permanent magnets in the flywheel to provide the field.  As such its inherently unregulated and has to have an electronic regulator that handles its full output current, that can react rapidly to load changes. 

Marine electronics in general is designed to run off the raw 13.8V supply that an older automotive type alternator + lead acid battery provides so your installation is less risky than a normal one.  However you may wish to add an over-voltage cutout circuit that disconnects the OB charging circuit if the battery is being over-charged to protect against regulator failure.

I'm still running a job lot of early 90's electronics on my boat, including an ICOM VHF and a Garmin GPS, which survived the previous owner's severe charging system problems and several alternator regulator failures.  That's 25 years without any extra protection against spikes and load dumps.

If you are feeling paranoid, a polyfuse feeding the equipment with a suitable high current TVS diode after it directly across the supply to clip any spikes, followed by a high current choke and a large electrolytic filter cap bypassed by a few uF of low ESR ceramic caps could be a worthwhile addition.

[Corporate666]

Use an RBO40.

It's made for exactly this purpose.  3 terminal device, supports up to 40 amps of current and handles overvoltage, reverse current protection and (most importantly) load dumps.

Just make sure you know the current consumption of your electronics and size any necessary heat sink appropriately.

[Ian.M]

    It seems to be a pretty widespread problem of spikes killing navionics, especially GPS modules. They're often wired to be on continuously. It sounds like the problem is commonly caused by spikes created by the starting motor coil creating a kickback of voltage from the inductance of the motor windings. I don't know why marine electronics would be more sensitive than automotive electronics. Though I'm sure regulations demand that automotive stuff be pretty bombproof. I wonder if the robustness of marine electronics hasn't kept up with the increasing sophistication and sensitivity, specifically with the now widely available consumer grade navionics. I'm sure the commercial stuff is way tougher- you're not going to be out in the Bering Strait with equipment that will crap out on you (I live in Seattle and you see a lot of heavy duty stuff down at Fisherman's Terminal.)
     In the consumer grade stuff damage caused by power issues is expressly not covered by warranty....

I think that spikes due to the starter motor are unlikely to be responsible, unless the electronics is *REALLY* badly designed, as the high voltage spike is on the motor side of the starter solenoid, and the solenoid coil side of the starter switch with no route back into the rest of the electrical system.  However if the equipment's regulator is badly designed and isn't protected against a sudden drop in the supply voltage (without disconnection), starting with a weak battery *WILL* kill the regulator.

Its far more likely that the issues are due to load dump transients from high current loads like electric windlasses, bow thrusters and 12V microwaves.

Use an RBO40.

It's made for exactly this purpose.  3 terminal device, supports up to 40 amps of current and handles overvoltage, reverse current protection and (most importantly) load dumps.

The RB040 will only survive a load dump if it has enough impedance upstream of it.  If fitted next to the battery without a polyfuse  or similar, its likely to fail shorted if a conventional fuse or MCB is used or catastrophically if unfused.

However a PM alternator (e.g. OB flywheel alternator) cant generate the usual type of load dump pulse when its used with an electronic series or shunt regulator as the regulator is much faster acting, resulting in a far narrower lower energy spike that's much easier to filter or clamp, and the RB040 will be suitable for the O.P's  application *IF* its clamping voltage is low enough for the (unspecified) electronics.

[Corporate666]

However a PM alternator (e.g. OB flywheel alternator) cant generate the usual type of load dump pulse when its used with an electronic series or shunt regulator as the regulator is much faster acting, resulting in a far narrower lower energy spike that's much easier to filter or clamp, and the RB040 will be suitable for the O.P's  application *IF* its clamping voltage is low enough for the (unspecified) electronics.

How many vehicles out there today *don't* use such a setup?  I've got tens (maybe a couple hundred) of thousands of devices out in the field and out of the hundreds we've gotten back over the years, I think there are maybe one or two failures that could have even potentially been attributed to load dumps - and that's on devices with less protection than something like an RBO40.

Design/engineering standards aside, we would have come out hugely further ahead just to skip a lot of the expensive protection against load dumps and just fix whatever affected parts came back.

Since it sounds like you design for automotive, I am very curious what your real world experience has been with load dumps - or are you designing for an OEM where you have to protect to a certain standard, regardless of the likelihood of the product experiencing what the standards protect from?

[Ian.M]

The marine market is *different*.  e.g. there is still a significant section of the sailboat market that still values engines that will continue to run after total electrical system failure.  That means no electronics on the engine, mechanical fuel pump and shutoff etc.  Then there are older boats - its not particularly uncommon to have 20 or 30 year old engines, well maintained and in good running order.  Therefore your chances of finding effective load dump protection already installed are much smaller.  Also its usual to have manual switches for selection of battery banks, which if operated incorrectly when the engine running, are an instant source of load dumps.

Add the expectation that navigation and communications electronics wlll be high reliability and the economics of protection vs replacement end up rather different.

Road vehicles with more than three wheels with PM alternators are vanishingly rare.  OTOH they are more or less standard on all except the largest motorbikes.

[Richard Head]

A motor vehicle has a low impedance ground (chassis) that effectively eliminates ground bounce.  Most small boats (and many small aircraft) are glassfibre and ground bounce can be a huge problem, mainly during starting. Make your star point the engine block and not the battery negative as there may be a substantial instantaneous voltage drop across the negative lead during cranking. In some instances if the battery is located far from the engine the negative lead could be ten or fifteen feet long.

[ccnw]

My original post may have looked like that buck/boost module was just a consideration, so there have a been a lot of great alternatives that are probably more "correct" spike protection. However, I have actually already have one of these in hand. It should work nicely to regulate the voltage for my electronics. So my original question remains, as to whether this module in itself would prevent spikes from getting through regardless of how they're generated. (If it self-sacrifices in the process, that's okay at the price...)

[Ian.M]

If I follow the chinglish, that module although rated for 5A max, is only rated for 3A continuous output current.   That would be pretty marginal if not overloaded when running a marine VHF and a plotter/fishfinder.  When transmitting at 25W, the VHF alone is likely to pull over 3A. e.g. my old ICOM M56 pulls 5.5A @ 13.8V when transmitting.

The *ONLY* way this would work would be if the average current of the VHF on receive + plotter/fishfinder is less than 3A and you can figure out the differences between the module you have and the similar one with a current limit pot and convert it to add a 3A current limit.  Then you could set it up to float charge a 12V 7AH SLA battery at 13.8V off the main battery, which would provide for the much higher transmit current and also give you a reserve for emergency comms if the main battery is flattened by binge poumps, cabin lights etc.

I suppose you could use it for just the plotter, but it is likely to be less spike resistant than the plotter is, and without some sort of clamping circuit or a battery on its output, you are betting that it will die quietly.  If instead it looses regulation as it dies, it could easily stuff over 20V into the plotter for quite a few seconds before its fuse blows.   Since you need the clamping circuit anyway, you might as well leave out the DC-DC converter . . .

IMHO, save it for something non-critical its actually good for like dimming cabin lights!

[ccnw]

Thanks, Ian. Yeah, I was thinking about the 5 amp draw on the VHF today and realized I'd hit the max. I like the idea of using the  module to just keep my SLA topped up (I have a 7 AH unit currently under the helm) and rely on the SLA for peak current on the VHF. But I hadn't thought of the possibility of the module going bad and slipping into the boost phase to create havoc. My outboard puts out 14.6V at cruising speed so I'm starting to think of using a buck converter- I just saw one that will display input and output voltage which would be handy. Sounds like I still need to come up with separate clamping circuit preceding it all anyway, so I'll look into previous suggestions.

[Ian.M]

14.6V is within the normal range of a 12V flooded cell Lead Acid battery charging system.  It is ever so slightly on the high side so you should check the acid level in your battery(s) regularly.  If  you are using a SLA (gelcell or AGM) battery, however, 14.6V is only OK for cyclic charging.  If you maintain that voltage once the battery is fully charged, you will shorten its life.   Depending on your engine usage that may not be a problem, but it needs 13.8V or so for long term float charging so if you run for hours at a time, you probably need a smarter regulator that has a float mode.   At a pinch, you can add a high current silicon diode in series with the regulator output to drop 0.6V or so with a switch across it.  If you know the battery is well charged and its a long run, keep the switch open.  If it was previously partially discharged, or for short runs close the switch.
I really don't advise a SLA battery as the main battery given your regulator's charging voltage - you should have a flooded cell leisure battery (or alternatively a dual purpose battery if the O.B.is electric start) with removable caps so you can top it up.

However there is no need to reduce the voltage for your marine electronics as it will be within their designed operating range.  The only equipment that would need it reduced would be domestic 12V equipment like a LCD TV not designed for portable use or domestic LED lighting.  Reducing it could even be counter-productive as a switched mode power supply circuit as found in nearly all modern equipment, that is within its design voltage range needs to draw more current to maintain the same output power if you reduce the input voltage.

Simply fit effective spike protection as discussed earlier, and a digital voltmeter on the instrument panel so you get some warning if the regulator fails and starts over-voltaging the battery so you  can limp home by either turning on enough lights to keep the voltage down, or by disconnecting the charging cable till the battery voltage has dropped below 12.5V

[DanielS]

If I was seriously concerned about load dumps damaging equipment, I would use isolated DC-DC converters so the input voltage would have no direct path to the loads. This way, even a catastrophic failure of the input filtering and clamping circuitry would have very little chance of affecting the loads.

[ccnw]

Thanks for the details, Ian.