@Zapper77 I'm already running some of the beta firmware - so I'm ok if you want to do an update on mine. With lower peak sunlight now I'm running my battery charge at 12kW (215 Adc) to ensure it gets as full as possible.
Last edited:
EG4 Flexboss overheating solved
- Thread starter alvitdk
- Start date
Even better, can you test it with your mod unit to see if it goes above 70°C @ AC Chrg > 9Kw ---> 12kW?
Zapper77
Renaissance Man (EG4 as of Jan 2025)
- Joined
- Jul 23, 2022
- Messages
- 2,759
Maximum Temperature on the t1 LV DC transformer is 125-132C (depending on model).
Is this a spec on the component(s)? 132C?
Or was this an actual test during AC charge and if so, at what pwr levl for how long?
since you're working on this, do you know if the 3rd pin on the fans in the FB21 are used by the control electronics? as far as I know it's a RPM sensor but it could also be a rotor lock sense signal. Either way, does the control electronics actively look at what's going on on those 3rd pins, i.e. would the unit shut down (or at least throw an error) if those pins aren't reporting what they're supposed to?
Zapper77
Renaissance Man (EG4 as of Jan 2025)
- Joined
- Jul 23, 2022
- Messages
- 2,759
Pin is RPM and the inverter uses it to detect locked rotor.
so hypothetically, if the fans are disconnected, the inverter will throw an error when it tries to turn on the fans? or would that just be a warning? what's the expected behavior in case of a rotor lock detection?
Zapper77
Renaissance Man (EG4 as of Jan 2025)
- Joined
- Jul 23, 2022
- Messages
- 2,759
According to the engineers, the potted inductor are derated when the t1 temperature exceeds the 125/132C threshold.
wpns
Solar Joules are catch and release
Can you confirm that? The last time we did a deep dive into the 18Kpv it turned out the third pin was called Locked Rotor, not RPM, and the two are different...
Also, do you have (PWM) control over the fan speed? Mine ramp up and down so quickly it's hard to tell, but it seems like they spend most of their time either full-on or full-off.
Zapper77
Renaissance Man (EG4 as of Jan 2025)
- Joined
- Jul 23, 2022
- Messages
- 2,759
The full on and off is completely changed on the new logic.
My 18Kpv fans stay on at idle once triggered for the majority of the day. (9kw PV, averaging between 5-8kw production) base house load 1,100-2,300 watts.
wpns
Solar Joules are catch and release
Normally yes, but I've gotta keep the lights on now that My Favorite Wife is on-island.
wpns
Solar Joules are catch and release
I mean are you able to feed the fans a variable voltage (generated by PWM in your circuitry) that will run the fans at less than full speed. It sounds like the answer is yes, but that's a feature of the new firmware, and has not been previously implemented?
It's not based on a set time.
I finally figured it out after having 2 of the oem fans disconnnected for months.
Inverter only sends a detect signal if temps reach >60°C. This is on t1. Haven't tested t2.
And, it correlates exactly to the app monitor's 4min sample rate. IOW, you won't know about it until it logs the sample in the app and can send out notification via email.
alvitdk
Solar Enthusiast
Not sure if it is needed (but always good to keep the temperature rather low than high). Either way, I have changed the design a bit to accomodate for the 12kpv as well. It works with the FlexBoss18 & 21, as well as 12kpv and 18kpvHis is pretty snazzy
I am working on a project with a 12kPV. Will it require extra cooling as well?
It could also pure voltage control but that limits how low the rpm can spin. On axial dc fans, the start up voltage can be anywhere from 5 to 9v usually around 7.
Modulating the 12v by applying the external duty cycle has been done but it’s usually avoided as modulating the electronics built into the fan is never good.
Really most dc fans without pwm control are usually meant to work continuously.
wpns
Solar Joules are catch and release
To be fair, until it tries to run the fan it won’t know if it’s not turning.
Thats with oem fans running, when either PV or Invert loads are >6kW but less than 7kW, temps are < 55°C (in conjunction with mod fans on top)
If getting near 9kW ( I have to AC charge to use that pwr for this 18) for some time, temps can get near 69°C and thats when the alert goes out. Under my normal operations, I never go >60°C.
I've testd this specifically to see if/when fans give the alert.
So the logic to check fans is based on temp @ >60°C, but logic to turn on fans are temp or pwr. Confusing and hard to keep track.
Quattrohead
100% off grid - EG4 Electronics
The beta fan logic is definitely an improvement but there is still more work to do, we are continuing to give feedback to the engineering team over in China.
For example I noticed the fan speed continually changing as the PV was going up and down with cloud cover, it would be better if there were some hysteresis built in to this so that the fans were not hunting, it's kind of annoying sounding.
We are working on it guys and it will get there, it takes time.
The great thing is that even though the original inverters are over 2 years old, we are still working on firmware improvements for them and really do take the feedback seriously, it just takes time to get the many different features improved. These machines serve a lot of purposes for a lot of people and they all have their own demands for improvements.
For example I noticed the fan speed continually changing as the PV was going up and down with cloud cover, it would be better if there were some hysteresis built in to this so that the fans were not hunting, it's kind of annoying sounding.
We are working on it guys and it will get there, it takes time.
The great thing is that even though the original inverters are over 2 years old, we are still working on firmware improvements for them and really do take the feedback seriously, it just takes time to get the many different features improved. These machines serve a lot of purposes for a lot of people and they all have their own demands for improvements.
wpns
Solar Joules are catch and release
Sorry guys, but until EG4 brings firmware development in-house with the proper resources we’re going to be stuck with “Hey, try this!” “Better, but you broke <something else>”, Rinse, Lather, Repeat.
There was some indication that they were going to do this, but it’s all gone radio silence from the EG4 software team. Too bad, I still maintain they could take the industry by storm. But here we are with the “same old same old” and the only advantage I see is higher prices.
There was some indication that they were going to do this, but it’s all gone radio silence from the EG4 software team. Too bad, I still maintain they could take the industry by storm. But here we are with the “same old same old” and the only advantage I see is higher prices.
Are you saying OP's design drops the temp 15C more than yours? Even with 1 less fan?
wpns
Solar Joules are catch and release
No, I’m saying:
And I had just installed mine that day so I didn’t know how it compared. This is the last few days:
Mine (Alpha):
And his (Beta):
So his actually looks better. BUT. Alpha has more PV and carries more of the AC load because <EG4has no idea why>. Now I need to swap the fans and let it run some more.
Oh I thought you had already tried it before that. My bad maybe I didn't follow the dates well on this thread.
Looking forward to seeing your post swap numbers.
What printer did you print yours on? I currently have a 220mm one at home which I'd rather use cause I have it super dialed in, but I'm not sure if that's too small. Not at home currently so can't try to put it on my slicer until I get back next week.
Last edited:
Would you be kind enough to add pwr ( inv and mppt) to those graphs?
From my data, AC charging leeds to the highest temps ( should be obvious since it's the least efficient), but I'd love to compare high power PV to Batts to see temps.
Similar threads
