150/180 Watt Electronic Dummy Load under-voltage fix

Peter_G

New member
Joined
Apr 21, 2019
Messages
1
Peter of HBPowerWall famerecently posted a video on theeBay150/180 Watt Electronic Dummy Loads,theyare great unitsapart from the under-voltage feature that's rubbish. To fix this Ive added a custom disconnect unit that uses a low voltage IO sense relay and a 12volt 100amp automotive relay (low contact resistance) to disconnect the battery until the unit is reset. It works like a charm and now my 100Ah, 360Wh, 40P1S battery pack capacity tests are very consistent. If anyone is interested in thisadd-on under-voltage unit please add a comment and I'll postthe circuit details. Cheers

image_flyqmu.jpg
 
Hello,
I got a similar dummy load too, with another pcb and smaller load-terminals but I think the other parts are mostly the same.
My unit hs a undervoltage feature, but there are two problems:
1. so far I couldn't figure out how to adjust the cut-off value (default is 2,0V)
2. the measuring is wrong. Without load activated the shown voltage is correct, but with load activated the shown voltage is round about 1V under the real voltage

Is there any option to solve these problems?
If not I would be very interested in your relais solution
 
Most of these cheap electronic loads seem to bevery low quality based upon reviews.The only good one I am aware of is the 150W 60V/10A load described here, which also does DC IR in the latest firmware, and has a PC interface.After afew years ofheavy use, my 2 units still work as new - withfactory calibration stillspot-on down to mV/mA. They'retruly a steal at $25.
 
^^^ Yes, you can set the discharge termination voltage to any value (with resolution 0.01V) and you can set a cutoff timer up to 100h (99h59m59s). Long press the upper right button to enter the settings menu, short press the low-right button to move between various settings, long-press low-right to exit settings. The up/low left button inc/dec values.

Iirc this youtube review (of a low-quality clone) has the same UI so you can see how it works there (ignore the remarks there about the deficiencies of the clone - which do not exist in the genuine units; and notethe genuines areonly known to be reliably available from one Aliexpress seller)
 
The problem is the DL24 load is using only 2-wire (vs. 4-wire / Kelvin) voltage measurement so it is reading the voltage at the end of the wires (at the load's input terminal) instead of the true voltage at the battery. This terminal voltage will be lower than the actual battery voltage (by amount I*R where R is the total resistance of the wires between the load and the battery).

The numbers in your video show that your pack resistance seems to be around 1.5 m? but the wires to the load seem to be around 30 times greater at 46 m?, so just the wires themselves account for a 0.69V drop at 15A.

As a workaround (in CC mode) you could adjust the cutoff voltage to be 0.69V lower than the actual target (but beware that the wire resistance will change as it heats up). But it's better to get a load with 4-wire support - like that I linked above. It makes little sense to design a load supporting large current but lacking 4-wire support. As I remarked before, the design of some of these cheap loads leaves much to be desired.
 
the used cables from the load to the pack are 2,5mm and are fitting the terminals at the load to 100%

I thought about this workaround too, or even to recalibrate the voltage measuring at the software menue,
but I don't think it's reliable.

Yesterday the dummy load stopped discharge-test after 1,5 hours at it's default 2,0V cut-off voltage, but with the multimeter I got 3,8v.
I want to use this device for testing, not for guessing.

So can you say that the linked dummy load has a reliable cut-off function?
 
^^^ Even if you used higher gauge cables there's still going to be a nontrivial voltage drop in the cables at high currents like 15-20A. Furtherit is unlikely that this load is actually properly designed to support such high currents. I recall reports of many of the clones dying quickly at the high end of claimed current support.

Yes, the discharge stop voltage ("sv") in theload I linked works exactly as desired in 4-wire mode, i.e. it stops discharging when the battery reaches the stop voltage as measured through the voltage sense wires (not the separate load wires) - see my answer in that thread.
 
I have been doing this stuff for years, and I have yet to find a better solution than a quality RC balance charger for these types of tasks.

Not only it is a 4 wire measurement, its fully aware of each cell group voltage, fully adjustable for any chemistry, ll while being cable of meting power in and out.
 
^^^ Yes, you can also use RC hobby chargers, but they are notorious for quickly dying if you use their internal loads much (even with higher end junsi iChargers). That may be less of a concern if you use them with external loads (or regenerative), but that's a bit of a pain to set up (and you can fry the charger if you are not careful). I've lost count of the number of complaints I've seen about chargers dying due to nontrivial use as discharger. It's better to put the wear and tear on a $25 load vs. a $250 charger.
 
^^ Good luck. Be sure to order from the known-genuine Aliexpress seller I linked there.

Btw, this youtube video shows the DL24 frying in only a few secs at 20V/8A (start at time 13:50). He says it blew a fuse and the IRFP260mosfet(which he thinks was fake). It's in Russian but you can turn on subtitles and auto-translation.

There are many similar reports of such failures on these clones. So don't believe the specs. It may not even handle 150W (which the genuine unit handles with ease).
 
gauss163 said:
^^^ Yes, you can also use RC hobby chargers, but they are notorious for quickly dying if you use their internal loads much (even with higher end junsi iChargers). That may be less of a concern if you use them with external loads (or regenerative), but that's a bit of a pain to set up (and you can fry the charger if you are not careful). I've lost count of the number of complaints I've seen about chargers dying due to nontrivial use as discharger. It's better to put the wear and tear on a $25 load vs. a $250 charger.

I have owned RC chargers for 15 years and I have yet to nuke one using within its capabilities. I find a lot of value being able to monitor the individual cell groups, which is why it is still my preferred tool.

Another solution I like a lot is the West Mountain Radio - CBA
http://www.westmountainradio.com/cba.php

I also like having nice tools. I don't mind spending on stuff like that.If you are budget constrained then sure... for $25 seems decent. That said, if you want cheap, you can just any load that has a constant draw and a stopwatch.
 
CrimpDaddy said:
I have owned RC chargers for 15 years and I have yet to nuke one using within its capabilities. I find a lot of value being able to monitor the individual cell groups, which is why it is still my preferred tool.

You're likely very lucky then. I've lost count of the number of such failures I've heard about. Most RC hobby chargers aren't properly (thermally) designed for discharging, so using them heavily for such will greatly decrease their lifespan.

Note that we can also monitor cell voltages using an electronic load and a BMS (or similar device).

CrimpDaddy said:
Another solution I like a lot is the West Mountain Radio - CBA
http://www.westmountainradio.com/cba.php

The WMR CBA has far poorer price/performance ratio than said$25 load. The cheapest version cost 7x more ($175), has less power (only 100W continuous), has fewer features, and is more cumbersome to use (requires a Windows PC to control it), to mention a few of many deficits.

The $25 load iscompletely programmable using a serial interface. So not only can we control it manually by its buttons, but we can also program it on almost any computer (e.g. an Arduino) to do all sorts of handy functions. Below are a few of infinitely many possibilities:

(1) DIYCC/CV charger: turn any normal (CV) power supply (e.g. wall warts, laptop bricks) into a CC/CV charger. Furthermore, we can completely control the charge algorithm, e.g. we can easily implement fast-charging algorithms such as multistage CC/CV algorithms used by Tesla and bleeding-edge cell phones (e.g. Qualcomm's 50% charge in 5 minutes). [If it's not clear how to combine a CV power-supply + CC discharger into a CC/CV charger then I can elaborate].

(2) Regenerative charging, i.e. use one battery as power source to CC/CVcharge another (same as (1) with a battery replacing the power supply)

(3) Balance charging (up to 60V/10A/150W), same as (2) but with balance as termination condition.

(4) CC/CV discharging: program discharges to have a final CV phase, e.g. to discharge to exact storage voltage, or to any target voltage (e.g. for determining SOC at exact voltage). Such discharging with a final CV phaseis the symmetric reflection of CC/CV charging (called reducing discharge "D-reduce" on the SkyRC MC3000).

(5) Variable load testing (to see how power sources respond to load changes, as in industry standard testing of power supplies and batteries).

(6) IR testing:we can programdifferent kinds of tests for internal resistance, both DC and AC.

The sky is the limit once we have complete computer control of the load - which is not possible with RC hobby chargers or the CBA (they are limited to a few built-in programs).

CrimpDaddy said:
I also like having nice tools. I don't mind spending on stuff like that [...]

If you like having nice (proper) tools then I highly recommend that you get a real (programmable) electronic load. For $25 the above load is a great deal (even a steal). A good load is an absolute must-have for any power user. The longer you procrastinate buying one the more time and money you will end up wasting (as I know too well from experience - both mine and colleagues).
 
I do have an electronic DC load... but still tend to gravitate to my iCharger 4010 Duo with an external resister load-bank. I also regularly use regenerative charging/discharging.

I do a lot of EV module testing and still prefer this as I can log cell group data in bulk without too much hassle and increased saftey since the charger has various ways to terminate the charge/discharge which is pretty important when you have to leave it unattended.
 
Stefanseiner said:
now I have another problem:
When load test is running the voltage detection of the tester is wrong.

at this weekend I changed the setup a little bit
- shortened the cables to 15cm
- using massive screw terminals instead of the thin crocos
- load test at 10A instead the maximum of 20A


image_krknao.jpg



image_gmsldc.jpg



image_nbkcmj.jpg


and I made a video of the discharge test so the seller can see the problem and Aliexpress too, because actually it's an opened case / dispute

image_iitgyp.jpg





but I guess you already know the result: the cut-off is still not working. Test stops at 2.80 volts but the battery pack has more then 3.8 volts left

image_tiuzay.jpg


Hidance already modiefied a test sample to 4-wire measuring and promised to send it out to me.
I am excited where they added the voltmeter cables, maybe I am able to do the same modifications at my actual load device.

image_zukqob.jpg


We will see if this package will arrive...
 
^^^ It stopped short of 2.8V because it was set on a 1:00 timed discharge. You need to turn that timer off.

But even with that fix it still won't take the battery down to 2.8V because of the approx 25m? resistance in your two-terminal leads, so the battery will instead be about 2.8 + 10(0.025) = 3.05V when the discharge stops (and it will then rebound to a higher voltage).

Workaround: to discharge pack to V, set cutoff voltage to V - 0.025*I, where I is the discharge current, e.g. use 2.55V above.

If you switch to another pack and/or different wires then you'd need to adjust 0.025 to the new wire resistance value.

Note that the wire resistance will change a little as it heats up, so the cutoff will not be exact, but it will be fairly close.
 
Stefanseiner hi can you take a picture of the soldering points and chips? i have the older model on the way :(
 
Back
Top