View Full Version : Daikin VRVIII Short of refrigerant?

31-08-2017, 06:34 PM
Hi guy's.

I have a system not working properly, but only had time for a brief look today with observations below. The client has asked that I don't do a reclaim / plt etc unless I'm certain it's short of refrigerant as the rooms are occupied by important clients and they don't want the system off unless necessary.

It's a master / slave multi unit set up (2 outdoor units with interlinked pipework)
There are two units (cassettes) in adjacent rooms which are not cooling very well at all. Around 4c diff between air on / air off and gas pipe TH sensor showing around 15c. The other rooms appear ok.

I tried shutting down some of the other units to see if more liquid would get through to these and improve performance, but there was not much difference?
When I had a quick look at the outdoor units, one had the condenser fan running all the time, but it seemed it was almost just moving ambient air, air off the unit seemed ambient although the compressor was running. The other unit was cycling the condenser fan, but the discharge air was nice and warm when the fan was on.
Suction line temp to both seemed high at 15c.
I noticed that the eev in the cassettes was only opening to circa 150 steps, so around 50%?
I'd have assumed they would be open further as the units were not hitting anywhere near setpoint?
I ran both of these in heating to see what happened and they both seemed fine? - nice hot air off with a good td.

Does this system sound like it may be sog to you guys, or is there anything else I can look into to clarify further?
I can't look at this again until Monday.

Thanks for any help - I'm a bit rusty on vrv's tbh.

31-08-2017, 06:48 PM
How do you know that the eev's are open 150 steps? Do you have a checker attached?
The max opening of the eev is 2000 pulses not 300 so at 150 it is nearly closed

31-08-2017, 08:32 PM
I'd have done the same, ran just the faulty units on their own to see what they do. Haven't got a checker so would have to do it the long way but would go something like:
Getting to the EEV and unscrewing the head to see if that makes a difference.
Try all units in heating to see what they do.
If it's a 3 pipe system then the changeover box has a 4wv in it. Crud idea really. Get a service manual and check the pipe diagram to figure the circuit. Have only ever seen them stick in heating. Yours would have to be stuck halfway so could rule that out for now I suppose.
Reclaim would probably take a couple of hours, perhaps not to bleed it dry but enough to tell if you are short or not.
If the customer thinks they are more important than they guy fixing the AC leave it on test heat a bit longer.


31-08-2017, 10:06 PM
Cheers guys.

The eev reading came from the wall controller. Others were working fine at 150 steps, but not these two. That's how the controller readout read..."eev 150 steps".

Re the client, I just do what I'm asked - I'd prefer just to recover the system and be done with it tbh.
The suction temp at the outdoor units seems high to me though - what's you're thoughts on that?
I'll try unscrewing the eev head when I return as suggested.
Gut instinct to me is sog - don't think it's a control issue.

Not sure about this, but apparently these can self diagnose a refrigerant over / undercharge?

31-08-2017, 11:23 PM
What about the thermisters/sensors? I have had this thing happen. Someone else ended up going out and reclaiming the gas. He weighed it out and it indicated no leaks. Found a problem with one of the thermisters/sensors, sorry can't remember which one it was.

01-09-2017, 08:13 PM
Hiya Paul,

All the thermistors appear to read correct!

04-09-2017, 04:07 PM
VRV is able to detect sog fault when its a big sog. VRV 3 should have a refrigerant charge check mode, but reclaiming the charge is the most reliable way.
The standard procedure to jugde if there is sog (near full load):
1. Сoarsely: if compressor discharge is over 100 C then its sog or major overload
2. More fine: Liquid subcooling before subcooler is about 7-12 K - OK, subcooling after subcooler not less than 20 K - OK
3. Superheat on the farthest indoor is 5-10 K - OK

14-09-2017, 06:26 PM
Thanks so far chaps, however the plot thickens.......

I found a leak in the condenser coil of the master unit. Reclaimed the entire system and 7kg was removed, and condenser coil was replaced with new.
The install docs showed the complete system charge of 23.5kg. This was weighed in and it was run through test mode.

The two units in question are still not cooling properly with poor td's over the coils.
Now, I also looked at another room on this system. It has two cassettes off one controller and one was performing better than the other.

It's as if the system is still sog??
Perhaps the install docs were wrong and it's had the wrong charge from day dot?
I'll try to measure the pipework tmrw, but much of it is hidden / poor access.

I'm wondering if it's worth trying the auto charge. I'd "assume" if the system still thinks it's sog, it will take in some more? - never used the auto charge myself.

Just to add;
Suction was 7 bar and discharge was 21 bar with all indoor units in cooling.
Suction felt normal - cold and wet.

What's your take on this??


14-09-2017, 08:17 PM
Have you checked the diversity of the system?

15-09-2017, 09:53 AM
Not always easy to tell just from the operating pressures as the system will try to self protect and adjust bypass valves etc to maintain the reference temperatures. The measurement of the pipe lengths needs to be a good estimate and not precise to the last mm. This should give you the best guide for the gas charge. I have never used the self charge as it can take quite a while to complete.
Are the four cassettes with poor td's at the end of the line? If so the system could still be sog.

15-09-2017, 07:32 PM
Don't forget the measurement rule....40 metres from first branch joint (refnet) to last fan coil.

Poor performance on fan coils at the end of the pipework run can indicate that the system does not conform to this rule.

You can exceed this rule though, providing that the pipe sizes have been increased in line with recommendations.

19-09-2017, 07:28 PM
Thanks for the replies.
Managed to measure the pipework and needed to add another 3kg. Original charge was incorrect from date of installation by the look of it. System appears to be working ok now.

As I said though, I'm a bit rusty on these, especially these heat recovery ones.
Another couple of questions if I may:

1. I set all indoor units to cooling, yet one of the outdoors had a nice hot condenser and good heat rejection. The slave however, had the fan running full tilt and the condenser was cold, as was the air off. Why would this be if every system was in cooling, IE none looking for heat?

2. Can I assume the room temp is controlled by the board looking at the return air temp in relation to the controller set point? Ie, the larger the difference the more it will open the eev?

3. In a couple of the rooms, the return air was 21c, the room was at 21, but the room temp according to the wall controller display was 24? Is there a field setting to offset the readout on the wall controller so it reads correct?

Cheers guys.

20-09-2017, 08:28 PM
Dont forget that the VRV III has options for high sensible cooling...


he new high sensible modes increase efficiency and comfort even further. In standard high sensible mode, the system uses an increased evaporation temperature, resulting in higher efficiency and higher comfort thanks to the high air discharge temperatures. The system can switch to high-power mode in the case of very high cooling loads, ensuring that a comfortable temperature can be reached in a minimum amount of time. Whereas in locked evaporation temperature mode, the high air discharge temperatures remain at all times, eliminating the risk of cold draughts. High sensible cooling modes not only increase efficiency and decrease unnecessary dehumidification, they also guarantee the absolute optimum in indoor climate.

21-10-2017, 11:34 AM
I think you may need a service checker on this system to diagnose further, VRV is exactly what it means 'variable' so it might use one HEX or two dependant on demand, ambient and head pressure, also might be worth going through field settings as a recommissioning process, with field settings i'd obtain service manual check which field settings relate to what you're trying to achieve, record what was set for each FCU, adjust accordingly and record again.