Testing Delta 2 efficiency under constand load

[Valaskjalf]

Hi guys,
So what is the best way to provide a constant load to my Delta 2 to accurately test efficiency? Im currently running a pc and screen through it, but the wattage fluctuates wildly between 75w and 105w so its hard to get an accurate estimate.

What can I use that produces a constant load without going the route of plugging in a super high-powered device?
Thanks!

 

[Tomtomtom]

Hi guys,
So what is the best way to provide a constant load to my Delta 2 to accurately test efficiency? Im currently running a pc and screen through it, but the wattage fluctuates wildly between 75w and 105w so its hard to get an accurate estimate.

What can I use that produces a constant load without going the route of plugging in a super high-powered device?
Thanks!

Use an external power meter plug (or smart plug with energy tracking) and you can get the average wattage over a few hours - then it doesn't have to be a constant load.

You need an external meter ideally anyway because you don't want to rely on the internal calibration of the device you are actually testing.

Otherwise tungsten or halogen lamps @ 50-100 W or so make great loads if you have any of those still lying around.

 

[Mirai]

Hi guys,
So what is the best way to provide a constant load to my Delta 2 to accurately test efficiency? Im currently running a pc and screen through it, but the wattage fluctuates wildly between 75w and 105w so its hard to get an accurate estimate.

What can I use that produces a constant load without going the route of plugging in a super high-powered device?
Thanks!

It's been tested already mate. It's in the low 80% s with some improvement if you add the extra pack by 3% and is better for AC > DC.

Although this guy got 74% efficiency from his:

 

[Valaskjalf]

Use an external power meter plug (or smart plug with energy tracking) and you can get the average wattage over a few hours - then it doesn't have to be a constant load.

You need an external meter ideally anyway because you don't want to rely on the internal calibration of the device you are actually testing.

Otherwise tungsten or halogen lamps @ 50-100 W or so make great loads if you have any of those still lying around.

Cool thanks will do!

 

[Mirai]

Use an external power meter plug (or smart plug with energy tracking) and you can get the average wattage over a few hours - then it doesn't have to be a constant load.

You need an external meter ideally anyway because you don't want to rely on the internal calibration of the device you are actually testing.

Otherwise tungsten or halogen lamps @ 50-100 W or so make great loads if you have any of those still lying around.

He can run it down to zero and test efficiency that way.

 

[Valaskjalf]

It's been tested already mate. It's in the low 80% s with some improvement if you add the extra pack by 3% and is better for AC > DC.

Although this guy got 74% efficiency from his:

Ja Ive been going through many of the videos and they vary a lot (not sure how it can be so inconsistent) but mine seems pretty low...maybe 70% so wanted to have a more accurate idea.

 

[Mirai]

Ja Ive been going through many of the videos and they vary a lot (not sure how it can be so inconsistent) but mine seems pretty low...maybe 70% so wanted to have a more accurate idea.

This guy got 91% for AC, 85% for DC. That's for the 220V UK model.

While this guy tested many of these portable batteries and got 74% AC/DC effiency. The highest he got was for any of these batteries was 91%. That's the American 120V one.

 

[Mirai]

Ja Ive been going through many of the videos and they vary a lot (not sure how it can be so inconsistent) but mine seems pretty low...maybe 70% so wanted to have a more accurate idea.

It's a bit of an overhyped Chinese product. The only value proposition is the 5 year warranty.

 

[Tomtomtom]

He can run it down to zero and test efficiency that way.

I'm not sure what exactly is meant by efficiency in this context but I wouldn't assume the battery capacity is a known quantity in the first place - so yes I would run the unit down to zero, measure the energy extracted (with an external meter on the output) and call that the "effective capacity" for a given load.

Then do the same with the meter on the input side for a recharge from zero to 100%.

The difference is the "end-to-end" loss. But you can't really isolate the inverter output and charging input losses unless you know the "raw" battery capacity, which would need to be tested directly.

 

[STORMERSFAN]

If u plug in 400 watts(tv/ps4/couple of light).. How long will it last.. Thinking of getting one..

 

[Mirai]

I'm not sure what exactly is meant by efficiency in this context but I wouldn't assume the battery capacity is a known quantity in the first place - so yes I would run the unit down to zero, measure the energy extracted (with an external meter on the output) and call that the "effective capacity" for a given load.

Then do the same with the meter on the input side for a recharge from zero to 100%.

The difference is the "end-to-end" loss. But you can't really isolate the inverter output and charging input losses unless you know the "raw" battery capacity, which would need to be tested directly.

We're making basic assumptions here and not writing up a scientific paper. One can be pedantic and speak of usable battery capacity and then inverter efficiency but one can also use shorthand as this is sealed unit which only comes in one capacity, 1024Wh. You can't change the battery. Best you can do is order a 1024Wh LFP extra battery or a 2000Wh LiNCM one.

 

[Mirai]

If u plug in 400 watts(tv/ps4/couple of light).. How long will it last.. Thinking of getting one..

I've seen estimates from 74% to 91% efficiency. So 1.9 to 2.3 hours. You can double that if you spend R13K on the extra 1024Wh battery. If you register your warranty Ecoflow gives you a R1000 off coupon, so you can get the battery for R1000 less. If you plug in some panels you can charge while using it.

 

[Tomtomtom]

One can be pedantic and speak of usable battery capacity and then inverter efficiency but one can also use shorthand as this is sealed unit which only comes in one capacity, 1024Wh. You can't change the battery.

My point is just that if you take the manufacturer's word for it, 1024 Wh, and also the manufacturer's display of power, say 100 W, then it's actually telling you nothing to observe that the load runs for 10 hours. It is certainly not telling you that you have 97% efficiency.

Because the manufacturer could actually have installed a 800 Wh battery and miscalibrated the display to show 25% higher power.

I don't think it's academic to want to confirm you got what you paid for. The only way to do this is to hook up an external power meter, and measure the energy drawn out in a full discharge. It's also useful to do this again after a few hundred cycles / months to confirm the deterioration is within spec.

 

[STORMERSFAN]

I've seen estimates from 74% to 91% efficiency. So 1.9 to 2.3 hours. You can double that if you spend R13K on the extra 1024Wh battery. If you register your warranty Ecoflow gives you a R1000 off coupon, so you can get the battery for R1000 less. If you plug in some panels you can charge while using it.

Damn that's alot of money for something ething that might only last 2 hours..

 

[Bundu]

I'm not sure what exactly is meant by efficiency in this context but I wouldn't assume the battery capacity is a known quantity in the first place - so yes I would run the unit down to zero, measure the energy extracted (with an external meter on the output) and call that the "effective capacity" for a given load.

Then do the same with the meter on the input side for a recharge from zero to 100%.

The difference is the "end-to-end" loss. But you can't really isolate the inverter output and charging input losses unless you know the "raw" battery capacity, which would need to be tested directly.

I agree.
For me only 2 things are important.
Usable full capacity and
Charge/discharge efficiency.

 

[Mirai]

My point is just that if you take the manufacturer's word for it, 1024 Wh, and also the manufacturer's display of power, say 100 W, then it's actually telling you nothing to observe that the load runs for 10 hours. It is certainly not telling you that you have 97% efficiency.

Because the manufacturer could actually have installed a 800 Wh battery and miscalibrated the display to show 25% higher power.

I don't think it's academic to want to confirm you got what you paid for. The only way to do this is to hook up an external power meter, and measure the energy drawn out in a full discharge. It's also useful to do this again after a few hundred cycles / months to confirm the deterioration is within spec.

Yes one can do that. My point was that you don't have to trust the display, you only have to run it down from 100% to 0. If you know the load you're running you can use a stopwatch. If you want to be more precise use a smart plug - although who calibrates those?

Low spec battery - that would imply fraud right? As this thing sells all over the world, someone would have caught them out on this. Secondly the thing can be weighed, no?

 

[Mirai]

Damn that's alot of money for something ething that might only last 2 hours..

Yes that's true. But if you look at the competition, it's not cheaper, and it's at best 2years of warranty.
The 2560Kw trolley someone bought in the other thread is a better value proposition. That's R25K (now 28K) and costs as much as the Delta 2 but gives same cycle count, 2560Wh and higher peak current. It's heavier though and seems to be louder. They only give 1-2 year warranty though.

 

[Tomtomtom]

Yes one can do that. My point was that you don't have to trust the display, you only have to run it down from 100% to 0. If you know the load you're running you can use a stopwatch. If you want to be more precise use a smart plug - although who calibrates those?

Yes I can see we're sliding into the wonderful world of metrology.

Low spec battery - that would imply fraud right? As this thing sells all over the world, someone would have caught them out on this. Secondly the thing can be weighed, no?

Of course and my example is extreme but the OP mentioned variability and this would be where the source of it would lie, probably, moreso than with the electronics.

Everything is built to a tolerance so what is the manufacturer's accepted range of capacity for 1024 Wh nominal? Is it 5% or 10% either side? Is 1024 Wh nominal not actually the centre of the distribution? It's still a bit of a wild west for these things.

 

[Mirai]

Yes I can see we're sliding into the wonderful world of metrology.



Of course and my example is extreme but the OP mentioned variability and this would be where the source of it would lie, probably, moreso than with the electronics.

Everything is built to a tolerance so what is the manufacturer's accepted range of capacity for 1024 Wh nominal? Is it 5% or 10% either side? Is 1024 Wh nominal not actually the centre of the distribution? It's still a bit of a wild west for these things.

OK. I suppose if one has precise requirements. Given one has a window to return these, eg if you buy in IC, you get 7-14days? You can buy it, test it with the stuff you will use it with, as STORMERSFAN has a usage scenario he wants to use this for, and if it fails to be useful, return for refund.

 

[KOPITE]

Every solution has it's pros and cons, the pps has the portability advantage to power certain stuff and specifically the ecoflow has the fast charging and user friendly updating of frequent firmware updates.

There is 2 important factors, price how much wh is required