Does anyone have any experience of protecting against surges on a 12V system? The problem I have is that in the winter my solar controller raises the absorption voltage level for charging the batteries, and then when a load switches off the brief time delay in the controller reducing the power from the panels causes the voltage to surge. It's not for long, but long enough to trip the inverter. The inverter is meant to be OK to 16V, but that's +/- 0.5V. If it trips while the fridge or freezer is running, when it comes back on that's not good for the compressor, and you can even get in a situation where the system can get stuck tripping repeatedly due to initial high power drain from the appliance trying to start up again.
At present I've limited the maximum charge voltage to 14.75V, which results in inverter trips being fairly rare, but means that in cold weather the battery does not get fully charged. So, I'm looking at alternative solutions...
There seem to be two types of surge protection available, one which is intended for temporary use on a car and can be picked up for around £50: http://www.sealey.co.uk/PLPageBuilder.a ... uctid=7237
The other is intended for marine use, and is about £120:
http://www.bepmarine.com/en/80-707-0004-00
The first of these doesn't actually say what it clamps the voltage to, the second one says 17V, which is going to be too high to be of real use in my situation.
The other solution is to change the inverter for one that has a wider voltage tolerance, but that's expensive and I like the one I already have in every other respect than this one.
Any ideas?
12V surge protection for solar PV
Moderator: Peak Moderation
-
mikepepler
- Site Admin
- Posts: 3096
- Joined: 24 Nov 2005, 11:09
- Location: Rye, UK
- Contact:
-
adam2
- Site Admin
- Posts: 10895
- Joined: 02 Jul 2007, 17:49
- Location: North Somerset, twinned with Atlantis
I fear that there be no simple answer to this one, I am afraid.
Most surge arrestors are intended to handle very short term spikes or surges such as might be caused by switching inductive loads or by faults, or by induced currents from external sources.
In this case, it seems to me that the problem exists for a second or more.
If the system is loaded, then the charge controller will put the maximum available charge into the system, especially when bulk charging the battery.
When the load drops, it takes perhaps a few seconds for the charge controller to "realise" and to start restricting the charge current so as to keep the battery voltage within normal limits.
The battery voltage then exceeds the maximum inverter input voltage and the inverter shuts down.
The battery voltage then drops and the inverter re starts, perhaps damaging fridges or similar equipment.
Any standard surge arrestor wont help in this case.
The only work around that I can see is to insert in the battery lead to the inverter a large silicon power diode that will drop about 0.7 to 0.8 of a volt and thus keep the input into the inverter within the acceptable range.
To avoid energy waste at lower voltages, the diode needs to be bypassed manually or preferably automatically at voltages below say 14 volts.
It will need to be a very large diode, on a very large heatsink. Or possibly several smaller diodes but that introduces complications to make sure that share the load.
Most surge arrestors are intended to handle very short term spikes or surges such as might be caused by switching inductive loads or by faults, or by induced currents from external sources.
In this case, it seems to me that the problem exists for a second or more.
If the system is loaded, then the charge controller will put the maximum available charge into the system, especially when bulk charging the battery.
When the load drops, it takes perhaps a few seconds for the charge controller to "realise" and to start restricting the charge current so as to keep the battery voltage within normal limits.
The battery voltage then exceeds the maximum inverter input voltage and the inverter shuts down.
The battery voltage then drops and the inverter re starts, perhaps damaging fridges or similar equipment.
Any standard surge arrestor wont help in this case.
The only work around that I can see is to insert in the battery lead to the inverter a large silicon power diode that will drop about 0.7 to 0.8 of a volt and thus keep the input into the inverter within the acceptable range.
To avoid energy waste at lower voltages, the diode needs to be bypassed manually or preferably automatically at voltages below say 14 volts.
It will need to be a very large diode, on a very large heatsink. Or possibly several smaller diodes but that introduces complications to make sure that share the load.
"Installers and owners of emergency diesels must assume that they will have to run for a week or more"
-
mikepepler
- Site Admin
- Posts: 3096
- Joined: 24 Nov 2005, 11:09
- Location: Rye, UK
- Contact:
The fridge uses about 45W running, but surges to at least 36oW starting up. The freezer is about 90W, though the surge seems not to be any higher than the fridge.
Adam, the delay for the controller to respond to the load dump is under a second, perhaps even close to half a second, but this is still too long for the inverter to be happy with.
The marine surge protector I linked above is supposed to be specifically to deal with this situation on boats (with power from a generator rather than PV), but as its cut-off is 17V I don't think it will help me.
I wondered about insulating the batteries - I never had this problem when I had them indoors, but I moved them into a ventilated outdoor shed to avoid risks from hydrogen build-up. But I doubt insulation would help much in a prolonged cold spell.
Adam, the delay for the controller to respond to the load dump is under a second, perhaps even close to half a second, but this is still too long for the inverter to be happy with.
The marine surge protector I linked above is supposed to be specifically to deal with this situation on boats (with power from a generator rather than PV), but as its cut-off is 17V I don't think it will help me.
I wondered about insulating the batteries - I never had this problem when I had them indoors, but I moved them into a ventilated outdoor shed to avoid risks from hydrogen build-up. But I doubt insulation would help much in a prolonged cold spell.
-
mikepepler
- Site Admin
- Posts: 3096
- Joined: 24 Nov 2005, 11:09
- Location: Rye, UK
- Contact:
It's a Mastervolt 2kW, capable of 4kW surge.
It's usually lightly loaded (fridge, freezer, laptop), but sometimes runs the washing machine (7-800W with heating turned off) or a kettle (650W), and when it's sunny I run a dessicant dehumidifier if there's nothing else to do with the power (350W).
I got it secondhand, so it was a reasonable price, though if I was buying another one day I'd probably look at a Victron as they have a wider voltage input range.
It's usually lightly loaded (fridge, freezer, laptop), but sometimes runs the washing machine (7-800W with heating turned off) or a kettle (650W), and when it's sunny I run a dessicant dehumidifier if there's nothing else to do with the power (350W).
I got it secondhand, so it was a reasonable price, though if I was buying another one day I'd probably look at a Victron as they have a wider voltage input range.
If it's the inverter that is disconnecting due to spikes, email the manufacturer [or any friendly equivalent maker] to ask if you can put a capacitor across, and close to, it's solar voltage input. I would guess it's fine, but I'm being cautious. They might suggest a value. What solar input voltage is it? If it's 100+ volts then maybe a motor run capacitor like this:
http://www.ebay.co.uk/itm/60uF-Motor-Ru ... lzLfV7siSA
Only buy from a real UK source [many are overseas posing as UK] and look for 5% tolerance and polypropylene as some sellers are vague and sell cheaper types as 'don't know the difference'.
http://www.ebay.co.uk/itm/60uF-Motor-Ru ... lzLfV7siSA
Only buy from a real UK source [many are overseas posing as UK] and look for 5% tolerance and polypropylene as some sellers are vague and sell cheaper types as 'don't know the difference'.
-
adam2
- Site Admin
- Posts: 10895
- Joined: 02 Jul 2007, 17:49
- Location: North Somerset, twinned with Atlantis
I do not feel that a capacitor would help in this instance, unless of truly improbable size.
The circumstances are IMHO, not truly a surge or a spike, but a prolonged over voltage, for a second or a significant fraction thereof. Surges are normally considered to be for less than a hundredth of a second.
Consider the following figures.
Normal charging voltage in winter-------15 volts (my estimate)
Inverter trip voltage, worst case---------15.5 volts (from o/p)
PV input in bright sun---------------------30 amps. (unknown, my guess)
Now connect a capacitor across the inverter input. Under normal winter charging conditions this capacitor will charge to the same voltage as the battery, 15 volts.
Now suppose that the load suddenly declines and that it takes the charge controller one second to react. During this one second, the capacitor has got to absorb the 30 amps solar input and in so doing prevent the voltage from rising to over 15.5 volts.
Since one amp for one second will charge a one farad capacitor by one volt, simple arithmetic shows that we need a 60 farad capacitor. Larger would be better to allow a margin of ignorance, perhaps 100 farads.
100 farad capacitors do exist in this voltage range but are very expensive, and remember that by slowing the rate of voltage increase, that this may slow the reaction of the charge controller.
Fitting such a capacitor would improve inverter performance under peak loads, but OTOH capacitors that big make me a bit nervous ! The actual energy content is not that great, but the speed with which it could be released is potentially dangerous.
The circumstances are IMHO, not truly a surge or a spike, but a prolonged over voltage, for a second or a significant fraction thereof. Surges are normally considered to be for less than a hundredth of a second.
Consider the following figures.
Normal charging voltage in winter-------15 volts (my estimate)
Inverter trip voltage, worst case---------15.5 volts (from o/p)
PV input in bright sun---------------------30 amps. (unknown, my guess)
Now connect a capacitor across the inverter input. Under normal winter charging conditions this capacitor will charge to the same voltage as the battery, 15 volts.
Now suppose that the load suddenly declines and that it takes the charge controller one second to react. During this one second, the capacitor has got to absorb the 30 amps solar input and in so doing prevent the voltage from rising to over 15.5 volts.
Since one amp for one second will charge a one farad capacitor by one volt, simple arithmetic shows that we need a 60 farad capacitor. Larger would be better to allow a margin of ignorance, perhaps 100 farads.
100 farad capacitors do exist in this voltage range but are very expensive, and remember that by slowing the rate of voltage increase, that this may slow the reaction of the charge controller.
Fitting such a capacitor would improve inverter performance under peak loads, but OTOH capacitors that big make me a bit nervous ! The actual energy content is not that great, but the speed with which it could be released is potentially dangerous.
"Installers and owners of emergency diesels must assume that they will have to run for a week or more"
Looking at your posts again, I take it you have solar charging 12V batteries through a controller, then batteries running the inverter from 12V? Again, I think you need a capacitor across and near the inverter 12V input because the transient from a sudden change is causing ringing in the wiring from the inverter back to the battery. If it's just a 12-16V average then maybe a big electrolytic? eg 22000uF £5:40 here:
http://www.ebay.co.uk/itm/Large-Electro ... rq6Qn5D7Ng
You can get screw terminals if you don't solder.
http://www.ebay.co.uk/itm/Large-Electro ... rq6Qn5D7Ng
You can get screw terminals if you don't solder.
My post overlapped with adams last post, so I will query the setup. If the charger is charging the batteries, then the batteries will swallow current with any long [second] rise in voltage? The batteries then supply the inverter with about 12-14V. My guess is that a sudden drop in inverter load [a fridge turning off] causes a fast rise in the input voltage to the inverter which has a 16V shutoff [strangely unhelpful spec]. This fast spike on the wire between battery and inverter is what the capacitor at the inverter input is swallowing.
Or is there something I am missing?
Or is there something I am missing?
-
mikepepler
- Site Admin
- Posts: 3096
- Joined: 24 Nov 2005, 11:09
- Location: Rye, UK
- Contact:
I'd considered a capacitor, but was wary for the reasons Adam mentions.
The battery should be charging to around 15V in winter, but even with the maximum target voltage capped at 14.75V, the controller itself records spikes as high as 15.23V - I suspect they may briefly go higher than this. It has a high-voltage-disconnect setting, but when I tried this it didn't disconnect, as I think the spike is too short for it to do so.
The controller can supply up to 60A, and because I'd be trying to keep things at 15V, I think it would need to be a very big capacitor to absorb enough charge over just a few tenths of a volt.
Still, as I can get a moderately sized capacitor fairly cheaply, it wouldn't be difficult to test it out as a temporary arrangement. I guess putting a fuse inline with it would make good sense as well, in case of a fault developing. If it made some kind of difference, then maybe I could lay out the money for a really big capacitor.
The battery should be charging to around 15V in winter, but even with the maximum target voltage capped at 14.75V, the controller itself records spikes as high as 15.23V - I suspect they may briefly go higher than this. It has a high-voltage-disconnect setting, but when I tried this it didn't disconnect, as I think the spike is too short for it to do so.
The controller can supply up to 60A, and because I'd be trying to keep things at 15V, I think it would need to be a very big capacitor to absorb enough charge over just a few tenths of a volt.
Still, as I can get a moderately sized capacitor fairly cheaply, it wouldn't be difficult to test it out as a temporary arrangement. I guess putting a fuse inline with it would make good sense as well, in case of a fault developing. If it made some kind of difference, then maybe I could lay out the money for a really big capacitor.
-
mikepepler
- Site Admin
- Posts: 3096
- Joined: 24 Nov 2005, 11:09
- Location: Rye, UK
- Contact:
-
mikepepler
- Site Admin
- Posts: 3096
- Joined: 24 Nov 2005, 11:09
- Location: Rye, UK
- Contact:
OK, one more interesting fact. As far as I'm aware, these surges DON'T trip my 180W Victron inverter (which runs solar water heating, router and laptop), despite it giving 15.5V as it's maximum. I've tried leaving it connected during eqalisation, and it certainly tripped then, as expected.
Presumably this is due to the Victron being more tolerant than the Mastervolt, or could something else be a factor?
I do have a spare 350W Victron inverter around, which is capable of running the the OR the freezer, so I could experiment using that for one of them and seeing if it is immune.
There's a further option actually:
- I keep the Mastervolt inverter just for running the big stuff, such as washing machine, iron, kettle. When this stuff is running there's no/little chance of an inverter trip anyway due to the power being used.
- I buy another Victron inverter to run the fridge and freezer, either the 800VA version which can take up to 17V input, or the newer VE.direct 500VA (this is nice as it's programmable)
details here:
https://www.victronenergy.com/inverters ... 0va-1200va
https://www.victronenergy.com/inverters ... 50va-500va
Presumably this is due to the Victron being more tolerant than the Mastervolt, or could something else be a factor?
I do have a spare 350W Victron inverter around, which is capable of running the the OR the freezer, so I could experiment using that for one of them and seeing if it is immune.
There's a further option actually:
- I keep the Mastervolt inverter just for running the big stuff, such as washing machine, iron, kettle. When this stuff is running there's no/little chance of an inverter trip anyway due to the power being used.
- I buy another Victron inverter to run the fridge and freezer, either the 800VA version which can take up to 17V input, or the newer VE.direct 500VA (this is nice as it's programmable)
details here:
https://www.victronenergy.com/inverters ... 0va-1200va
https://www.victronenergy.com/inverters ... 50va-500va
How big is the Mastervolt inverter, and does it have an 'always on' load? I've seen some ~3kW inverters that consume 20-30W even when there's no load?
I've been thinking one solution would be to have something around 500-800W on all the time for lighting, IT, fridge etc. and a 2nd inverter around 3-5kW which is only switched on when needed for the big stuff like induction hob, power tools etc.
I've been thinking one solution would be to have something around 500-800W on all the time for lighting, IT, fridge etc. and a 2nd inverter around 3-5kW which is only switched on when needed for the big stuff like induction hob, power tools etc.
Is the battery close to the inverter? In my theory it is the length of wire from battery to inverter [it's not coiled is it, or close to something magnetic?] that is the mischief although I have no idea if a 12v battery is inductive anyway. There is a manual here:
http://productimageserver.com/literatur ... 7545OM.pdf
Is that one of your models? It just seems weird that it should dislike over 16Volts input if batteries can be ~14V but it does say so.
http://productimageserver.com/literatur ... 7545OM.pdf
Is that one of your models? It just seems weird that it should dislike over 16Volts input if batteries can be ~14V but it does say so.