contactor vs. shunt trip breaker

[jtvt]

I was wondering what the consensus is for either of these for off grid.

Pros for contactor
cheaper
no hands on to close contactor

Cons for contactor
Uses power
need precharge

Pros for Shunt trip
no precharge required
doesn't use power for breaker to remain closed

Cons for Shunt trip
Need hands on to turn breaker back on


Am I missing something?

I'm not sure the hands on pros/cons mean much since if either disconnects, you lose power (networks) and you would have to visit the battery for both.

 

[toms]

Reviving this thread that received no input previously.

I’m about to put together a Batrium based system and am keen to hear reasons people used shunt trip breaker in preference to a contactor.

 

[400bird]

I went with a shunt trip breaker, I don't like the idea of wasting power just to keep the contactor closed 99.99% of the time.

For a vehicle application, where it will open frequently and possibly be open more than closed, a contactor is the logical choice.


The inverter needs a precharge either way, the contractor/breaker isn't likely to care.

 

[OffGridInTheCity]

Cons for Shunt trip
Need hands on to turn breaker back on

Some will point out (and I agree) that this is a Pro rather than a Con. If it trips, it can be good operational procedure to require manual restart - e.g. you're more likely to correct the issue.

-------------
I have Batrium and a 400a SACE S5 shunt-trip - mainly because e-bay/used ones are nicely priced for a 400a 'breaker'!

One advantage is that my SACE has a 24vdc coil but my battery is 48v. However, since it's only a 'pulse' I can tap my 48v battery at 24v and it won't unbalance the battery because it's a micro-second pulse when it trips. This eases battery -> coil matching in some cases.

 

[rhino]

I’m about to put together a Batrium based system and am keen to hear reasons people used shunt trip breaker in preference to a contactor.

I'm also preparing same thing and will be trying to integrate a REC recharge device into the system like this shows: https://diysolarforum.com/threads/batrium-bms-setup-8s3p.13952/page-2#post-297335

 

[toms]

I have installed many REC-BMS systems using contactors and the REC precharge module, i don’t have any experience with the use of a shunt trip breaker.

Specifically i’m concerned with the durability of the breaker contacts if they are disconnected under load, it seems that while they have a DC rating they are primarily used in AC applications.

There are a lot of Batrium systems using shunt trips, i’ve never seen another system use this disconnect method.

I’m wondering what the compelling reason to use a shunt trip breaker is?

 

[400bird]

I have installed many REC-BMS systems using contactors and the REC precharge module, i don’t have any experience with the use of a shunt trip breaker.

Specifically i’m concerned with the durability of the breaker contacts if they are disconnected under load, it seems that while they have a DC rating they are primarily used in AC applications.

Breakers have spec sheets that will tell you their ratings. Typically, they are rated to break the full load thousands of times. ABB for example has proven itself a highly reliable manufacturer for these breakers, and they are DC rated into the 100's of volts.

Here's the ad version from Midnite solar on their breakers:

MidNite's breakers are rated to break the full rated load at the rated voltage repeatedly, with NO DAMAGE. Always use a properly sized breaker for disconnecting

Product - MidNite Solar

MidNite Solar Product Page.

www.midnitesolar.com

Where as I've seen multiple contactors fail after opening under load dozens of times (dozens of openings causing a single failure and I've seen that failure mode dozens of times)

There are a lot of Batrium systems using shunt trips, i’ve never seen another system use this disconnect method.

I’m wondering what the compelling reason to use a shunt trip breaker is?

Reliability, no power use to stay closed, ease of control, designed, tested, and certified for exactly this use case.

 

[toms]

Thanks for that. I guess that leaves the only reason to use a contactor being the ability to reconnect remotely.

I will delve deeper into the shunt trip breaker rabbit hole !

 

[toms]

Some will point out (and I agree) that this is a Pro rather than a Con. If it trips, it can be good operational procedure to require manual restart - e.g. you're more likely to correct the issue.

-------------
I have Batrium and a 400a SACE S5 shunt-trip - mainly because e-bay/used ones are nicely priced for a 400a 'breaker'!

One advantage is that my SACE has a 24vdc coil but my battery is 48v. However, since it's only a 'pulse' I can tap my 48v battery at 24v and it won't unbalance the battery because it's a micro-second pulse when it trips. This eases battery -> coil matching in some cases.

Do you use any other fuse in conjunction with that breaker? It appears from the specs that the breaking capacity is 36kA, which is pretty impressive.

 

[OffGridInTheCity]

Do you use any other fuse in conjunction with that breaker? It appears from the specs that the breaking capacity is 36kA, which is pretty impressive.

In my case, it's an 18650 bank where each cell is fused. The purpose of the breaker (shunt-trip) is
1) To disconnect all battery current from the system if an individual pack of the battery is over charged or under charged or too hot or too cold - to avoid damage to the battery till it can be fixed.
2) To manually disconnect the battery to work on things.

The reason ABB SACE is popular is they specialize in HI AMP DC disconnects and there are many 'used ones' on ebay which brings the price within reason. https://www.ebay.com/sch/i.html?_fr...acat=0&LH_TitleDesc=0&_odkw=SACE+S3&_osacat=0
As mentioned above, a 400a disconnect can have arcing and other issues - so its a real science and you want a proper breaker.

Note: I have down stream breakers - e.g. each 12,000w inverter has it's own 250a breaker. And in fact I tripped on once. But even though 'in theory' the 2 inverters could reach 500a it doesn't happen in practice so the 400a overall shunt trip saved me a few $ instead of going with 500a. Also the battery can handle up to 600a - so the battery is well protected from overall load. Here's a visual:

 

[OffGridInTheCity]

There are a lot of Batrium systems using shunt trips, i’ve never seen another system use this disconnect method.
I’m wondering what the compelling reason to use a shunt trip breaker is?

Because Batrium is designed to provide 'a pulse' for a shunt-trip. For example, I have the "Critical Fault" event associated with Relay #1 and this sends a pulse when the system detects a Critical Fault. Its does not send a continuous current as you would need for a regular relay or contactor - you'd have to figure something out.

 

[jtvt]

I'm the OP.

I went with a Shunt trip. Those ABB units are solid.

I think a compelling reason for a contactor might be a situation where the battery disconnects frequently. However, I would assume the more disconnects, the more likely the contactor will die or fuse at some point. Probably even more so at high amps.

That use case is different than the off grid use case where a disconnect should never happen. If it does, you likely have to visit the battery. That was my final determination. If there is a problem, I have to visit the battery because a BMS triggered disconnect is going to be a cell failure, temperature problem, or hardware failure. All other cases get handled by the inverter or cc. I lose power or get a warning before the battery gets to a BMS level problem.

IMO, the precharge is partly there to help the contactor and with the correct equipment that's all it does. IIRC, Schneider was doing something to limit inrush on their XWs. One could also argue that if inrush is a problem, you need a bigger bank. I haven't had a problem that would suggest a need for a precharge.

One thing that is limited on a Batrium is you can't have multiple independent contactors or shunt trips on different strings. The batrium will monitor multiple strings (really as just one long string), but it can't control an independent disconnect per string. It doesn't do per string monitoring such that each string can be shutdown independent of others. I personally wouldn't go there anyway, but you need multiple batriums if you want multiple disconnects.

 

[OffGridInTheCity]

One thing that is limited on a Batrium is you can't have multiple independent contactors or shunt trips on different strings. The batrium will monitor multiple strings (really as just one long string), but it can't control an independent disconnect per string. It doesn't do per string monitoring such that each string can be shutdown independent of others. I personally wouldn't go there anyway, but you need multiple batriums if you want multiple disconnects.

You're correct about this. Batrium allows you to trigger external relays on a long list of options BUT, you can only do 1 option per relay. Critical Fault is the main 'overall battery error' option and you can only assign this to 1 relay.

 

[rhino]

You're correct about this. Batrium allows you to trigger external relays on a long list of options BUT, you can only do 1 option per relay. Critical Fault is the main 'overall battery error' option and you can only assign this to 1 relay.

This seems really bad.. What's the point of having the WatchMon CORE and bunch of K9's for multiple strings if you can only ever turn off the entire bank? If you have two strings with a K9 each one I would want the K9 to be able to turn on relay for the string that K9 is connected to. You are saying that isn't possible? If so then would think it would be better to just use something like two JK BMSs.

 

[OffGridInTheCity]

This seems really bad.. What's the point of having the WatchMon CORE and bunch of K9's for multiple strings if you can only ever turn off the entire bank? If you have two strings with a K9 each one I would want the K9 to be able to turn on relay for the string that K9 is connected to. You are saying that isn't possible? If so then would think it would be better to just use something like two JK BMSs.

This is beauty in the eye of the beholder kind of thing.

To me - the purpose of Batrium/shunt-trip is *protection* of the entire battery - and figure out what's wrong. I don't see an advantage of turning off 50% or 25% of a battery bank and expose the remaining part to overload as an automatic process.

I would point out that one could do dumb breakers per parallel battery to manually disconnect a sub-set of a battery to work on it or to take a portion offline once the BMS alerts you to a problem. In my case I don't even need breakers, I can just unscrew a bolt to disconnect from the bus.

And, I would agree that alternate BMS designs are a perfectly valid choice. Batrium is just one type of approach.

 

[rhino]

I was thinking more along the lines if each string in a battery could easily handle the load on its own for the day and then extra strings is just for extending Wh's for cloudy days then having a string disconnected would be fine. Just wondering why Batrium wouldn't at least give you option to have a "critical fault" per K9 unless they are worried people wouldn't know to make sure the rest of the string(s) could handle the load properly.

 

[OffGridInTheCity]

I was thinking more along the lines if each string in a battery could easily handle the load on its own for the day and then extra strings is just for extending Wh's for cloudy days then having a string disconnected would be fine. Just wondering why Batrium wouldn't at least give you option to have a "critical fault" per K9 unless they are worried people wouldn't know to make sure the rest of the string(s) could handle the load properly.

Batrium is oriented to manage a single 'battery bank' and they strive to interface with a wide range of equipment which also is designed along this line. To do multiple 'battery banks' one would have to have multiple Batriums. It's just a choice they made.

The big advantage for me - is over the last 3 years I've added, and added, and added but still use the same 'core' + longmons. This year I plan to add batteries 7 and 8 taking me up to 112 packs (250 is the max). To add/extend my batter bank, all I have to do is daisy change in the additional packs and make a software setting....


Easy to view packs 'overall' and expand....


While I agree it seems like they could allow multiple relays to be associated with the same event - maybe they aren't sure how to change the software? But personally, I don't see this as a practical design goal.

 

[jtvt]

This seems really bad.. What's the point of having the WatchMon CORE and bunch of K9's for multiple strings if you can only ever turn off the entire bank? If you have two strings with a K9 each one I would want the K9 to be able to turn on relay for the string that K9 is connected to. You are saying that isn't possible? If so then would think it would be better to just use something like two JK BMSs.

I think the opposite. If one string is disconnected, then you have a long process to reintroduce that string. if you have multiple strings get toggled on/off for whatever reasons, the strings will be out of sync. You could very well have hugely different voltages when trying to automatically reintroduce a string. You could also drop enough strings that you no longer have sufficient power to start your inverter. IMO dropping a string is more of a PITA, then disconnecting the entire battery.

BTW, Batrium will do this with multiple cores.

 

[rhino]

When you re-introduce the string wouldn't the other strings simply transfer energy into or out of the introduced string fairly quickly to get them back in sync?

 

[OffGridInTheCity]

When you re-introduce the string wouldn't the other strings simply transfer energy into or out of the introduced string fairly quickly to get them back in sync?

In my case, a string is 14s100p ~13.5kwh easily capable of 100-200a. If you tried to hook in a 3.6v 'string' when the rest are at 3.9v - the current rush would likely be a problem.

I'm 18650. My system cut-off at 49.5v = 3.54v/cell in the night. This bounces up to 3.6v in the morning (7am) When I add new packs I discharge them to 3.6v before adding - and plan the work 1st thing in the morning with maybe a 100mv max difference. Personally, I would not try to add in when there a multi-volt difference such as 3.6v to 3.9v or 4.0v without calculating the current / wire size. In addition, that large of difference would take a loooooong time to stabilize at 13.5kwh battery level.

I don't have experience with LifePo4 but it's a flatter voltage curve and so there's a wider range of charge where the voltage is not that different - but that implies a balance problem that won't be visible till you try to charge them up nearer 100%. Maybe someone knows what happens here and can comment.

So the answer is - it depends