question

Hi Jeroen,
maybe it would be a solution to bypass the delayed Victron regulation, to consider battery consumption in order to control the triacs.

Hi Thomas ,

I do not know what you mean?

The Quattro's raise their frequency to reduce the solar inverters , how do I by-pass this when I need to use frequency to control the warm water triacs , not possible and also not wanted.

The Quattro's nor the GX´s (CCGX or Cerbo) have an analoge output which is programmable , I asked that many years ago already on the old forum.

When the frequency goes up (above 50.10Hz) the PID will send set the triacs further open to keep 50.10-50.15Hz by heating coils in water , so that the solar inverters do not reduce power.

When our large 3KW water pump stops this frequency jumps up to 52.5Hz and takes sometimes up to a minute to go back to 50.00Hz again , when this pump starts there is no frequency change.

When our small 2KW water pump stops it jumps to 51.8Hz , etc.

At both situations the triacs are set further open (sometimes to max 6KW) since the PID thinks it has surplus power from the solar inverters.

I've put a day timer (heating system on from 1100-1900) so it will not happen at night nor low radiation , and a switch on delay (90 seconds) to not react at all on this frequency spike from the quattro's above 50.05Hz.

Then I've put also a switch off delay (10 minutes) for cloudy days , otherwise the frequency is very high as we wait for 90 seconds switch on delay for nothing , and loose energy production , and then when it switches on it will go to 6KW for nothing , this is tested and clouds pass within these 10 minutes where I live.

In the solar inverters I've put a ramp down/up time of 30% per Hz , starting at 50.30Hz to 52.50Hz maximum , faster or slower will have a worse result , smaller or bigger dead band have a worse result , is tested already.

The assistant in the "Master" Quattro is set at 50.10Hz to 52.50Hz maximum , to give a dead band so the PID triacs can control the power instead of the solar inverters.

Can someone tell me how to stop this frequency raise by stopping a big load in the first place?

Is this maybe part of the the boost setting on the "Inverter" page in VE config , I thought this was only for the start of a big load (current booster) , or is this doing something with the frequency as well?

Jeroen.

WWW.QuintaCasadaFonte.com , North Portugal , check on VRM-world , no pass required.

Hi guys and ladies ,

I see I have a bunch of followers.

As can be seen in the other topics about frequency shift , the engineers face a challenge for that as well , there it stays high for hours and also during the night , connected to Lithium ESS systems.

I tested the current boost at 0.3 (lowest) for a couple of days , no result , then at 1.0 , no result , then at 3.5 (highest) , no result , default setting 2.0 , no result , still the frequency goes up very fast and goes down very slow after.

I discovered that it is even worse when switching the net at the input off through the aux input setting in an assistant so the Quattro´s have time to prepare so to say , the net goes of in about 10-30mS , then it jumps to 52.7Hz as well (within 0.2 seconds (200mS) , measured with a RMS Fluke) and then gradually goes down to 50Hz in about a minute.

It does this with or without sun , with load or without , at night , always .....

So I installed two Peter 5.5KW (800 euro each) soft starter-brakers on my water pumps , no result , still the quattro´s will give a small kick for the last 600-700W or so when it switches off on the minimum current from the E-motors , so , money thrown through the drain.

Well , please Victron ... can you make in the assistants which use the frequency shift (start at 50.2 to 52.7 and off at 53Hz standard data) a ramp-up and ramp-down setting , e.g. power% against seconds , %/S , OR seconds against frequency , S/Hz, that would help I think already a lot.

0 %/S would be off , no power reaction , 1 %/S would be 100% power reduction in 100 seconds , 40 %/S would be 100% power reduction in 2.5 seconds , 100 %/S would be 100% power reduction in 1 second , etc .....

OR , 0 S/Hz would be 50Hz (60Hz) constant , 1 S/Hz would be up from 50.2 to 52.7Hz in 2.5 seconds and is 100% power reduction , 40 S/Hz would be up from 50.2 to 52.7Hz in 100 seconds and is 100% power reduction , etc ....

In frequency shift enabled solar inverters and Lithium BMS´s (battery controllers) you can set that as well , why not in a Multi or Quattro?

These BMS´s tell the Quattro (Multi) when the battery is full , these can best be charged with a constant current 25A (example Pylontech 2.4KWh@48V) , meaning 1200Watt an hour , theoretically these are full in 2 hours , oeps , the sun is shining way longer than that and not at a constant power as well so a lot of excess energy will be thrown away by raising this frequency to limit the solar inverters.

In the Multi and Quattro there is a ramp-down time of about 0.04-0.05Hz/second since it takes a minute to come down.

In the solar inverters I've put a ramp down/up time (power reduction) of 30% per Hz , starting at 50.30Hz to 52.50Hz maximum as I mentioned before , meaning the power is not completely reduced at 52.5Hz , only 66% , so 34% of solar power is still going to my net when the Quattro´s want to reduce maximum , this 34% is then still going to my hot water coils , not to loose to much power , but you see that this process takes just 2,2 seconds , not 1 minute needed at all for a solar inverter nor a BMS!

Then , still my biggest question is , can we be sure when something big switches OFF or the Quattro´s (Multi´s) switching the net OFF these will not raise the frequency? OR it can do that with programmable ramp-up and ramp-down times in there so we could set these as desired?

Another small problem with this frequency shift is also that all clocks will run faster for a while , every month I have to reset my clocks (kitchen equipment , week timers , etc) about 10-15 minutes , and I run only 50.1Hz maximum during full battery and full sun , imagine you do not have these PID heating coils or a Fronius Ohm-pilot and you run day in and out between 50.2-52.7Hz , that´s about 1 hour ahead a month on all your clocks and a lot of energy through the drain!

I save about 1/6th (1.5MWh) on top of my total (9-10MWh) electricity production by using the frequency shift as an input for my PID coils , saving me again 350-400 euro a year.

Total cost for 2 phases to make it was just 300 euro (almost 5 years ago) , for a normal house it would be around 200 euro only , Fronius will set you back 650 euro for the pilot and another 100-150 for a net-metering unit to get about the same result but the pay-back time for a normal house is 10-15 years , too long , this one works off-grid as well and is back in the pocket in 4 years , Fronius only on-grid , you need 1 frequency to mA converter and 1 frequency switch , 0.02% or better , 1 PID controller , 0.01% or better , power chopper(s) or controlable SSR(s) depending how much phases you have , heating coil(s) , breaker(s) , some power relays , 2 timers , a day timer , a power meter , and a display to see actual power in % being fed into the water , enjoy ......

The other very small box from phase 2 has only a chopper , a breaker and a power meter and gets the same signal from the PID controller.

With the best regards , Jeroen.

Good morning gents and ladies ,

We have been testing some more.

We had this frequency going up before , some 7-8 years we remembered.

Resetting a WOLF solar pump controller every time it was doing that but our computers stayed online.

This controller starts the pumps from 8 Vacuum tube panels to transfer the heat to 2 accumulators of 1000 liter each.

After having been in contact with WOLF and a software upgrade it still happened.

This controller works only with normally open contacts and they close to start an event , e.g. start a pump , open a valve , etc.

We mounted S10K250 Varistors over every contact and the problem was solved till now , it never resets anymore.

So , to come back on the frequency shift , we now mounted over all relays (contacts) and the 2 soft starters (in and output) these S10K250 varistors , this seems to help since we do not have any more that the Quattro´s raise their frequency when the pumps switch off.

Can it be that when that when a contact opens (a little spark and/or a little back feed) and the Quattro/Multi that it will raise the frequency?

Is the back feed doing this maybe , for a fraction of a second this raises the voltage from 230 to 400+V , measured with a true RMS Fluke , recorded within 200mS.

The position of the Quattro´s in relation with the soft starters and pumps is 150 meters , that might have been the problem.

Well , that part is solved , still the remaining is when it switches off the grid.

Since that relay is somewhere in the Quattro , we do not want to open this of course , we will do a test to put first S10K420 varistors on the inputs and outputs of the quattro , between Neutral and the Phase.

If that is not working we will put S07K350 ones there , which have a little less resistance.

We will keep you updated.

With the best regards , Jeroen.

Good morning Gents and ladies ,

We have done the test with the varistors on the input (between N and L) and output (between N and L) of the Quattro´s , we have result.

The frequency is now "jumping" way less with them but it is not completely gone yet.

We tested some days ago with and without these varistors.

So , that means that the quattro´s are doing this when it probably senses a higher voltage when opening the input relay , switching off the net.

The idea is now to make a small box and to put these Varistors nicely there.

We will also do another test by putting these varistors also from L1-input to L1-output , N-input to N-output , etc , so that these are really set over the relay contacts completely.

That is normally done directly over the contacts from relays to reduce the "spark" but as mentioned before we do not want to open the Quattro´s to mount those inside.

Hopefully it will reduce the frequency jumping even more.

We keep you informed , Jeroen.

Good morning gents and ladies ,

Here I am again with the results.

We have placed the varistors as mentioned before , it helps a bit , but not all the jumping is gone.

Verdict , still , this must be a setting in the software.

But , as said before , that is something Victron should make possible to adjust in a setting which up to now is not available.

Another thing we tested is ; when the frequency is high , above the 51Hz , and then you switch to the net , black-out , then the Quattro is so quick to switch on the net , within 2 seconds , that it is not lowering the frequency to the frequency of the net.

When we just switch one fase 120 degrees to the wrong side , to see if that helps , yes it does , then it takes about 5-6 seconds as the Quattro has to line up whit the 2nd fase.

Then it switches without a black-out since also the frequency is then way lower and almost 50Hz.

But that of course is not normal to make the fase rotation on purpose wrong to not have a black-out , since it takes 5-6 seconds to line up , in case of an overload and it has to switch on the net for power assist , that takes to long and gives a black-out as well.

Well , we are out of possibilities to test , I hope somebody from Victron might shine light on this phenomenon.

With the best regards , Jeroen.

WWW.QuintaCasadaFonte.com , North Portugal , check on VRM-world , no pass required.

Hi friends

when searching for solutions to my (the same) inverter-frequency problem I tried to contact the Victron staff directly, but did not get any answer; then I found this thread and many, many users who seem to fight with the Victron's inverter' frequency laziness. It is desperating! As it should not be too complicated to reprogram the firmware with an assistant that linearily rises and decreases frequency by voltage change I don't understand the blocking behaviour of Victron at all.

Has anyone found a tricky solution in between? Or can explain why Victron does not care?

Martin Krahn

I think you will find uts not just one thing and thus victron don't like changing it as one change affects other stuff.

Ie

1. If the inverter/charger gets hot because it's at full charge rate and ambient temp us high. It needs to reduce the solar output. Thus Hz rises for no aparate reason. I found this happening as with gui mods installed on the gx it says on the screen temp so it soon out.

2. I am also using a assistant to open and close the ac 1 in relay. With a time delay on the close. I do find that if the close trigger happens then while the timer runs down I get a freq increase upto 53hz. I can't see a reason for why though. Maybe a coding error don't know.

Anyway there is more to the frequency change assusrants than just dc volts. I have the same errors when using the pv assistant as well as the essential assistant. But Only when I am also using general flag for ac1 in ignor. And assistants to turn ac1 in back on. Something there us amiss.

Maybe it's also affecting other stuff.

But yes its very annoying.