Miscellaneous > Logged in diaries

An Proto-Teach Éireannach

<< < (2/119) > >>

Scruff:
Let me introduce Wendy Wu's Pot of Gold. Named after the lovely lady that sold me these genuine CALB LiFePO4 raw cells and after a month of negotiation didn't drop her price but did give me 9 free bus links and the fancy plastic mounting armour. Apparently genuine sellers are hard to find, I shopped around she was the only one who had them. Pretty good deal.



Blimey that's a lottov switchgear! What am I compensating for?
MasterVolt....I'll explain later. Over-priced lemon is the jist.



That's only the middle layer.  :o

Next task whack her inna case!



Ta-Da!

Ahhh...wee problem...that's a waterproof case mate!



Not anymore!
Does that affect my warranty?



I bet you can guess where this is going...

It's a good way to make exhausts vents & intakes. Drill a bazillion holes. It's the lowest impact on case integrity I could think of. Guess how long that took... :-[

Scruff:
I'd like to shine a light on some myths that are getting passed as truths.

Lead can compete with lithium if weight and or space isn't worth money.
Lead has a non-linear efficiency. Operationally it can be >95% efficient to 95% SOC. Then it bottom dives which makes a 100% cycle ~80% efficient or 120% returned.
Most on the market chargers only charge to 90% or 95% (use one..then check your specific gravity)
You do not have to fully charge lead once a week unless you are using one of these chargers.
I charge lead to 120% once a month and I have 6-year-old batteries that are testing like new from a live-aboard with this regime because I don't use conventional chargers.

Let's have a closer look at a lead system. Lead-fi has been cycle tested.
A few notes first.
This is a piece of industrial hardware.
I'm rating it as 1kWh to the load.
I am not saying it's a 116Ah battery but only use half of it.
It's a 58Ah battery; use all of it (and the computer won't let you use more).
Cycle it to 95% in 3 hours & 20 mins and give it back to me when you are done or you'll be waiting until the cows come home for it to finish charging. I have a battery hospital for maintenance charging, that is not a user problem.

This is Lead-Fi's Mk-I battery



That was made of e-waste, it was underwhelming so I replaced it with mostly Yucells and relocated the fuses.

Plated capacity
8 x 7Ah Yuasa 12V +
4 x 24Ah Yuasa 12V
= 76Ah @ 24v

Actual capacity
116Ah!! @ 24V  8)

C20 Discharge test to 50% DOD, result x 2 = actual capacity.

116 / 20 = 5.8A Discharge Rate




Inverter Output



134.5W

139.5W in 134.5W out = 96% efficient inverter @ 10% Duty


Discharge time



10 hours 41 minutes.

End of Discharge @



58Ah removed



times two = 116Ah battery. Wow...I've ran several tests, every single battery in there is casually under-rated.

Inverter Output C20 (Elapsed Power)



1110W  Power to Load
~1400W extracted from the battery

79% System load efficiency @ 10% duty including temperature deratings and active proportional case cooling.

Scruff:
Now let's look at the battery and charger.



Charger Max Load


PSU Power In:



PSU Power Out:



90% efficient PSU @ 100% duty
99% efficient Charge Controller

Not bad...but not linear

Efficiency (and power factor) drops with Duty Cycle

High Absorption
80% SOC



87.5% Efficient PSU
98.6% Efficient Charge Controller






95% SOC





83.6% Efficient PSU
98.6% Efficient Charge Controller

At 95% SOC Charger has used



and returned 57.9Ah including cooling
hence to 95% SOC this lead is 95% efficient including about 1.5Ah of cooling power.

71% round trip efficiency (to 95% SOC).
Li-Fi was 60% round trip at C1.5 but then again the cables were 60°C.

I'll post the 100% result tomorrow after it arrives hours and hours later.

Scruff:
The charger was still absorbing at 130% SOC 24 hours later...I ended the test without checking the kWh counter. Sometimes MS controllers have difficulty delatching when you feed them power supplies, I may have a closer look at that.
It's not an issue for flooded, if you water them. I always wonder how mang bubbles an AGM has before expiry.

As a side-note for all you non-Irish speakers thinking it's a pretentious thread title. "Teach" means house as Gaelic, pronounced Tee-yach. I don't teach other than anecdotally.

Scruff:
I'm running a final round of tests on Lead-Fi to dial in the best curve LVD (voltage + time) in the absence of SOC based alarm assistance.
That Bep meter...junk...it reads too much noise to set a zero. It triggers low voltage alarms on batteries that aren't connected (0V) and the alarm is not a contact but a 2hz square wave...I'm amazed how many engineers don't use their own products.  ::)

That cheapo meter is much better...it's no high-class instrument don't get me wrong but it's very good value for money, the calibration is reasonable and no faddling required. Turns out the voltage is accurate, it's not lying every time I've validated it since.

So I left yee on a cliff hanger. Lead-acid efficiency, what does tinternet say? 80% was it?....right....

I've told my meter Lead-Fi's battery is 58Ah (116Ah 50% usable).



I removed 52.2Ah at C10



The best charger algorithm in production replaced 54.5Ah



Hrmm that's a 96% efficient battery*.  This is not unusual.



*including active case cooling.

Navigation

[0] Message Index

[#] Next page

[*] Previous page