Actually ST requirements are 100mA for lighting and 30mA for plug points (power points for equipment easily accessible by hand). In this case, 100mA main RCDs are acceptable, provided that you have 30mA to cover the plug points in addition to the main RCD. If there is only main RCD it has to be a maximum of 30mA, but ST doesn't really recommend 1 RCD for the whole DB.

Of course, this is in addition to water heater circuits which are required to be protected by 10mA RCD (dedicated).
Adding a 10mA RCD in cascade with a 30mA/100mA is not an issue. The device is meant to trip the power due to safety reasons, what may happen is that both may trip at the same time if there is a problem, which is more preferable to it not tripping when the problem occurs.

Doing it "flat" is of course preferable but will involve alot of work at the DB, which is dangerous if not done properly.
Both may still trip, due to tolerances of the device itself especially when 10mA and 30mA are quite close in rating. There is a proper way to do discrimination of tripping for the devices, but even without the discrimination it's still preferable to put them in cascade rather than not having sufficient protection.

Hi Kytz, thanks a lot for your reply.

For my RCDs, my electrician just daisy chain them from one RCD input to another RCD input. Ended up quite a long chain cause I've multiple RCDs. Is that dangerous?
 



As for discriminate tripping, I think this one really need proper sifu. I think 99% street sifu don't know this. Mine also don't know.

Physical installation of electrical circuits can get quite misleading. If the cables are terminated in a daisy chain as per your picture, that is a parallel(flat) installation instead of cascading. Because the looping is happening at input of RCD only.

For example in below case the RCD are separate and will only trip 1 RCD if a fault happens at A / B / C / D respectively.


But if cascaded like below any fault in B may trip both RCD, but a fault at A will only trip the bottom RCD.


It's considered cascaded when output of the RCD is connected to the input of the next RCD, as you can see in the picture above.

In your case, if all 4 RCD are installed in a cascade, it's not dangerous but likely to be redundant because any fault at the end of the RCD will likely trip all of them, unless there is sufficient discrimination between them. RCDs are protective devices, so they don't introduce danger through their operation. But if RCD is not installed or is faulty, then the lack of protection from the RCD can be a hazard instead.

Yes indeed you're right. Mine is linked parallel. Not cascade. Which I thought might be dangerous cause if the 10ma RCD failed means no more protection. So was wondering if I should change it to cascade.

Actually devices that can cause safety hazards like water heaters already come with built in RCD nowadays, so it's not necessary to cascade multiple RCDs (as long as all DB circuits are properly protected already with the correct ratings).
Rather, it's recommended (and also more important) to test the RCD regularly instead to ensure that protection is intact. Some water heaters even allow you to test their RCD.
Because like what ceo684 has mentioned, the RCD will fail someday, and the only way to find out is by testing them.

yes  , one can try use 30mA  RCCB at the DB box. 
the issue will be  more power trips .  as in Msia  we do have  many lightning storms .  and if you are not home
the food in the fridge will go bad.

the thing about 30mA for 13 A SSO   really  how many house  implement this . even in todays   new houses majority
dont have.
so how did they pass the tnb inspection /CF   and is it really necessary .  i dont really know.

Actually devices that can cause safety hazards like water heaters already come with built in RCD nowadays, so it's not necessary to cascade multiple RCDs (as long as all DB circuits are properly protected already with the correct ratings).
Rather, it's recommended (and also more important) to test the RCD regularly instead to ensure that protection is intact. Some water heaters even allow you to test their RCD.
Because like what ceo684 has mentioned, the RCD will fail someday, and the only way to find out is by testing them.

might be easier to add single 10ma rcd (with cover) next to heater switch outside bathroom.

yes  , one can try use 30mA  RCCB at the DB box. 
the issue will be  more power trips .  as in Msia  we do have  many lightning storms .  and if you are not home
the food in the fridge will go bad.

the thing about 30mA for 13 A SSO  really  how many house  implement this . even in todays  new houses majority
dont have.
so how did they pass the tnb inspection /CF  and is it really necessary .  i dont really know.

Hi ceo. I've got a question on point 3 of yours. How do you add the 10ma RCD under the 30ma/100ma RCD? my electrician told me if I do a hierarchy level that will cause side effects. eg. they trip each other off. so he told me to install flat level instead. eg. the rcds are totally isolated.

they  will not trip each other off.

example  if  there is leakage of 20mA  in the WH  ,  only the 10 mA  RCD  will trip.

Hi Kytz, thanks a lot for your reply.

For my RCDs, my electrician just daisy chain them from one RCD input to another RCD input. Ended up quite a long chain cause I've multiple RCDs. Is that dangerous?




As for discriminate tripping, I think this one really need proper sifu. I think 99% street sifu don't know this. Mine also don't know.

Actually devices that can cause safety hazards like water heaters already come with built in RCD nowadays, so it's not necessary to cascade multiple RCDs (as long as all DB circuits are properly protected already with the correct ratings).
Rather, it's recommended (and also more important) to test the RCD regularly instead to ensure that protection is intact. Some water heaters even allow you to test their RCD.
Because like what ceo684 has mentioned, the RCD will fail someday, and the only way to find out is by testing them.

Actually not a problem.. 30mA built in modern day with anti nuisance trip trips very little
Correct. It should be tested as part of every new year/CNY routine maintenance.
This is a plan B solution, but for ease of troubleshooting it is safer to put everything at the main DB.

Not many - its common in retrofitted and new build in SG
But in MY.. new builds got put but mainly China brands to save cost
Rarely you will get Schneider nowadays even for half million dollar place
30mA RCD, whole house (sum of all L - sum of all N return = acceptable losses)
Branches out from here to a bunch of little MCBs.
The 10mA is just for WH's own single L and single N. Input=output = acceptable losses and won't trip. IF the WH N is shared that means:
1) Installation is wrong in the first place
2) Useless for purpose of RCD, forever trip.
3) Most likely because it is not a dedicated LNE run from DB box to the WH unit.

Impossible they will trip each other off unless the WH circuit is shared neutral.
Correct.
And if there's a leakage of 20mA in a socket or light outlet unrelated to WH circuit, the 10mA will not trip.
No, this is acceptable.
Mana ada WH with built-in RCD.
ELCB got lah.

RCD is god tier protection (think of it as modern airbags)
ELCB is antiquated retired era stuff because "what if leakage don't leak nicely through Earth" aka leak elsewhere e.g. humans = don't t

Not many - its common in retrofitted and new build in SG
But in MY.. new builds got put but mainly China brands to save cost
Rarely you will get Schneider nowadays even for half million dollar place
30mA RCD, whole house (sum of all L - sum of all N return = acceptable losses)
Branches out from here to a bunch of little MCBs.
The 10mA is just for WH's own single L and single N. Input=output = acceptable losses and won't trip. IF the WH N is shared that means:
1) Installation is wrong in the first place
2) Useless for purpose of RCD, forever trip.
3) Most likely because it is not a dedicated LNE run from DB box to the WH unit.

Impossible they will trip each other off unless the WH circuit is shared neutral.
Correct.
And if there's a leakage of 20mA in a socket or light outlet unrelated to WH circuit, the 10mA will not trip.
No, this is acceptable.
Mana ada WH with built-in RCD.
ELCB got lah.

RCD is god tier protection (think of it as modern airbags)
ELCB is antiquated retired era stuff because "what if leakage don't leak nicely through Earth" aka leak elsewhere e.g. humans = don't t

Not missing but busy.

I apologize for not replying.The updates

1.Sunday no technician.So I suggest to increase the heater to 75% hotter ( basic 50% before it get less hot ) and spray into the water tub.I confirm that the water is not hot and I forgot to mention , the water flow is also slow.
2.Monday call 3 technician , all 3 said no need to put extra RCCB outside since all new or old water heater have Elcb build in.All 3 technician said something stuck cause it to get water not hot.
3.Tuesday , Call again to main joven and panasonic technician for repair but need to book the time next week and the cs said go to electrical shop and choose a model
4.Yesterday plan to go out and have a look for the model , heavy rain , didn't go out.
5.Today , see whether still rain or not at the evening. as what Momo33 said , buy water heater first , then only find a good technician.
All technician also didn't recommend to have an extra RCCB , i really don't know why as they said all buid in, no need anything extra.

As some technician told us better change but some said can fix.Most of them doesn't have cert or just normal electrical technician.
I also don't know which decision to choose where the water heater is less hot and water flow very slow.If water flow slowly should be much hotter.
Anyone know what is the reason of part failure when water heater not getting hot ?

Now finding a good water heater but budget not over rm400.
Any suggestion for DH-3MS1 and DH-3N series ?

Thanks , will reply to you all but I don't really understand much on those DB and model and brand.I need to research more into it.SOrry for not updating in time and thank you again

Looks like you have all advises you need on water heater. I was away from PC for at least 2 weeks now 

On old water heater not being hot, very likely that there is calcium carbonate build-up inside the copper piping at the heating elements. This will cause heat transfer to the water less efficient and waste of energy. I'm assuming that you are comparing the heat setting when the heater was new compare to how it is performing now. It is good idea just to buy new water heater.

Like one said, when incoming water is colder than usual like during rainy season, you do find that you need to turn the knob higher to get more heat. 

Last week i call to those electrical man , not as details as yours but they did said something was stuck and said can be fix and clean.

1st , I don't think it's because of rainy season , it's more like it's not hot.Example yesterday was fine at 50% tuner knot but today 75% only achieve the same temperature as 50%.It's 25% difference.
It's comparing yesterday and today.Just 1 day the difference it too wide.

I haven't go and check the piping but I am sure the water flow speed is same from yesterday and today and only the changes is the temperature need to set to 75% compare to yesterday 50%.
If the water flow flask is faster , I think need 99% to get 50% temperature.

What could be the reason ?

Last week i call to those electrical man , not as details as yours but they did said something was stuck and said can be fix and clean.

1st , I don't think it's because of rainy season , it's more like it's not hot.Example yesterday was fine at 50% tuner knot but today 75% only achieve the same temperature as 50%.It's 25% difference.
It's comparing yesterday and today.Just 1 day the difference it too wide.

I haven't go and check the piping but I am sure the water flow speed is same from yesterday and today and only the changes is the temperature need to set to 75% compare to yesterday 50%.
If the water flow flask is faster , I think need 99% to get 50% temperature.

What could be the reason ?

Possible the temperature control knob spoilt.

Now that you mentioned it is intermittent, then it is most likely that the potentiometer (variable resistor) that control the temperature setting is bad. Over time the brush contact and the carbon rail gets bad giving this problem just like amplifier volume control. I'm quite certain an old (even new one) is using this part as setting control. This is not expensive part, probably around RM10 (so long never buy this  )

If the piping stuck, the water flow would be slow, thus water gets hotter not cooler.