Greetings,

I wanted to test the integrity of some large scale batteries that are used in products to jump start vehicles. I am not a fan of battery load testers, I'm not a car enthusiast but it has been my experience that these testers do not draw the Amperage capable of even a small car thus NEVER accurate. I'm guessing for a small scale vehicle I would need to have the capability of drawing AT least 90Amps. Now I had two ideas and would appreciate input. They are as follows:

The first is if I was to purchase a car starter, will the motor draw the appropriate amperage, or does it have to be hooked up to the crank shaft in order to draw the 80 amps? (as I said I'm not a car person so if I mentioned the wrong components I apologize in advance)

The second idea is I was going to purchase two 1kW .25ohm resistors , having two in parallel, which would be about 100A's (I had to compromise a bit for pricing of the resistors based on ability).

I like my second idea better (maybe I could test it for a few seconds with a load meter and just watch and see what the battery outputs). My question is what do you guys think? I know it may seem overkill but those battery load testers (that cost like 40 bucks) never seem to work and I'd like to get as 'real world' test results as possible.

Thanks,

Chalma

 

I am both a car and off road truck guy .. But personal I take my batteries to advance auto or auto zone and have them tested for free.. As for building a tester I wouldn't waste the time or money cause you can buy a digital tester for about $40 or so and they really work....
What's the end result of this project ???
I ask cause I use jump packs all the time since I work for a Tow Truck Company and AAA... Generally these packs don't last long unless they are always fully charged and kept this way and not left with a partially charge... I just buy new batteries from Interest Batteries for about $50 bucks or ... Also most modern chargers chargers have some kind of battery tester built in and takes up less room ... Solar makes a really nice battey tester and harbor freight has a nice one ...

 

I would say you are on the right thought path to proper battery testing by making a high amp load bank system.

That said there are a lot of ways to make a high capacity load ballast resistance systems for far less than what a pair of .25 ohm 1 KW resistors will cost you.

One is the simple spiral wire used for common notebooks. Surprisingly its typically either a chrome plated steel or a stainless steel wire to which effect either takes a considerable amount of abuse as a low ohm high wattage resistor.

The second option is to do a bit of online shopping and pick yourself up a role of Nichrome heating element wire of which ever gage and lengths will give you the closest resistance and wattage capacity that you are designing for.

From there design wise figure the basic handheld medium current battery testers typically pull around 100 amps and do a test for 10 - 20 seconds to get a reasonable voltage drop rate and the bigger commercial ones are usually 200 - 400 amp load systems.

Personally if it was me I would be aiming for a load bank design that will do at least 3/4's of a batteries cranking amps rating for at least 15 - 20 seconds without overheat issues.

 

Wow thanks for the great replies.

@Dr.Killjoy The end result of my project is for work, we have exactly what you describe as jumpstart batteries for emergency situations. Once in a while we get returns, or I need to have someone test the the batteries every so often. These are a lot of units sometimes so I can't take them to another business. I will acquire the tester you mentioned though and test it out. heck it's not that much money (plus it's not me paying for it =p )

@tcmtech I had no idea I could purchase the Nichrome wire by itself. The prices of some of those resistors is outrages. I will see if I can get some and maybe dissapate it around some ceramic material. 15-20seconds is very helpful and I didn't think about that, I don't know why I thought instantaneous would be sufficient.

Thanks again guys, at least I know I'm heading in the right direction.

 

How about using some 12V 100W halogen bulbs in parallel. Each bulb draws a little over 8A and I've seem them on fleabay for less than $3 each. It would also give a direct visual indication of the battery condition.

 

Where I work we have a older analog style commercial battery tester that does its test at around 200 amps for 10 - 20 seconds to determine if a battery is good or not and so far the guys in the shop swear by its readings.

Now our fancy and rather expensive digital one that has at best 14 ga leads and takes maybe half a second to do its tests has let far too many bad batteries go out and cause trouble that anyone cares to admit to.

I just did test on our service trucks batteries last week or so with both testers.
The fancy digital one said both batteries tested at 90% or better whereas the old high load long duration tester put both batteries right at the bottom line of being acceptable at 20 seconds and fail at 25 seconds.

I am not sure how the digital units test a battery but I know for a fact that in real life engine starting watching the voltage drop on a high amp draw for any period of time will always tell you if a battery is good or bad.

 

The first is if I was to purchase a car starter, will the motor draw the appropriate amperage, or does it have to be hooked up to the crank shaft in order to draw the 80 amps? (as I said I'm not a car person so if I mentioned the wrong components I apologize in advance)

Hi Chalma, welcome to the forum. In order to properly load test a battery, it must be done at 1/2 CCA (cold cranking amps) for 15 seconds. This simulates an engine cranking over for 15 seconds and should be performed at roughly 70-80 degrees fahrenheit. When you are finished the test, you should have roughly 9.5V or better, depending on temperature. You must also have a battery with an open circuit voltage of 12.4V minimum before you start or the battery is destined for failure. So to answer your first thought, a starter without any load on it probably will not give you anywhere near the load you need to properly test the batteries. A heavy truck starter will only draw around 200 amps on a no-load test and automotive around 50-100 amps. You could however use the vehicle starter and perform the same test (as a load tester) with proper results.
As for your second idea, I don't feel that you can create the proper draw with your setup and even if you do, you are going to create a large amount of heat, and I think your components are going to be shortlived.
Proper testing must be done with a carbon pile load tester and you can get a cheap version of these for a couple hundred dollars.
The second tester you speak of is a conductance or electrochemical impedance spectroscopy tester and it sends a signal or frequency into the battery and measures the difference between the input and the output. I have been doing this for 35 years and in my mind, the only true test is the first one because it applies a true load to the battery internal components, connections etc and shows a true value. The second method is good for selling batteries, ask your local quicklube joints. Here is a cheap version of what I am referring to and it will give you real results. Good luck to you.

 

...
One is the simple spiral wire used for common notebooks. Surprisingly its typically either a chrome plated steel or a stainless steel wire to which effect either takes a considerable amount of abuse as a low ohm high wattage resistor.
...

Awesome! I love learning that very simple cheap household devices can be used to replace expensive specialty parts!

I assume you screw-connect the condustors to it? Soldering can't be much use.

 

a starter motor wont draw much by itsself, it has to have the load of an engine to draw full current.

 

I have and use one of these.

It works fine for occasional testing. The only drawback is that you have to wait a while before doing a subsequent test...

 

Thanks all for the wonderful replies and supportive comments. I have decided after much thought that I will go with Crutschows idea. I utilize a test like this on UPS/portable power supplies we also make using strands of 100+W bulbs and I don't know why it didn't occur to me to do the same with the 12V issue. I am also doing this for the safety of whoever will be testing these units. I can just imagine for whatever reason the people we have testing these units get a hankering or will to disregard safety and just touch these coils and get burnt...

 

Hi Chalma,
Two quick things. First of all, I never added the link which I was going to add at the end of my last post but I see MikeML picked up the ball for me and added exactly what I was going to put on there. No better way to test a battery.
As for your method you have chosen, a couple of questions for you. After giving you the proper method to test batteries, how do you plan on regulating the current draw on the batteries and how do you know when you have had a sufficient load?
Secondly, do halogen bulbs not throw off an immense amount of heat? Sorry, I'm confused

 

Secondly, do halogen bulbs not throw off an immense amount of heat? Sorry, I'm confused

Yes, but they are purpose-built to do just that while giving visual feedback at the same time. ANY load will generate the same amount of heat - that's where the battery energy HAS to end up. Some folks use things like immersion heater elements as dummy loads, since they are also built to handle large amounts of heat.

 

Note that a bank of 12 volt lamps will have a much lower resistance with a cold start, then increase as the filaments heat up.

 

Yes, but they are purpose-built to do just that while giving visual feedback at the same time. ANY load will generate the same amount of heat - that's where the battery energy HAS to end up. Some folks use things like immersion heater elements as dummy loads, since they are also built to handle large amounts of heat.

My point is that there is a specific way to load test vehicle batteries which I laid out earlier. I am not sure how a bank of lights is going to tell you how much load is being drawn and how much time you need to apply it to get the same results. Yes lights give a nice visual that there is a load, and I understand that electrical energy gets changed to heat and light, but this is not the solution for this problem. I find it redundant that MikeML and myself have given the OP a fairly inexpensive alternative which is a proper proven method, and everyone wants to reinvent the light bulb (pun intended). A carbon pile load tester is the best method, as I said earlier, and it is very accurate.

 

Similar thoughts here as well. I cant see how having a bank of lights or a inverter and normal household voltage bulbs is going to be simpler cheaper to build and all around superior to good old fashioned heavy gauge resistance wire and some sort mass that soaks up the heat without getting to a dangerously high temperature.


Especially if the OP wants simple, cheap, safe and portable to work with.

 

When ever these "load" threads come up, I have to mention my supply of 1Ω & .5Ω wire-wound-on-porcelain in the flea market.

Also have access to some absolutely huge load banks. Some in metal enclosure. At 7-Rivers prices.

If posts like this are not appropriate, moderators can delete.

 

Vehicle batteries are rated at the CCA ... Cold Cranking Ampere. As I recall, that is a 30 second standard.

A defective battery won't need the full CCA to indicate a failure. Applying 0.1C load can cause the voltage to drop below the low cell/battery voltage.

I've seen 12V 25 AH rated batteries drop below 10.75V in less than a couple of seconds with a 50 ohm load. That's not anywhere near a 0.1C (2.5 A). I've seen batteries that previously tested OK fail under full load. I used a 1kW water heater element to test both UPS systems and 24V batteries. The resistive load worked great in both conditions ... drawing about 2 amperes from the 24 volt source and 8 amps from the UPS.

I'd use whatever light bulbs that would create the best load for the system. The 1157 (turn signal and stop lamp) draws 2.1A and 6.8A respectively. A few of them in parallel and you can have a very good load bank for 12V systems.

 

........................... I am not sure how a bank of lights is going to tell you how much load is being drawn and how much time you need to apply it to get the same results. Yes lights give a nice visual that there is a load, and I understand that electrical energy gets changed to heat and light, but this is not the solution for this problem..............................

If you know the voltage and power rating of the bulb than you know how much current it is drawing. Of course its resistance is non-linear with voltage so it will draw more current at a lower voltage than a carbon pile (which means the current varies less as the battery voltage drops compared to a linear resistance, but that could be considered an advantage). They also can have a 100% duty-cycle without concern about overheating the elements since most of the energy is radiated.

Bulbs are indeed one solution to the problem. It's a matter of trade-offs and preferences as to which solution is best.

 

Okay I give up. Apparently whatever I have been telling my students and what I have known to be true for 30 some odd years, is just nonsense. Let's go with the light bulbs and wires hanging all over the place because that sounds a lot safer than a carbon pile load tester with a true history and impeccable track record. Did I mention the fact that you can buy one of these for under a hundred bucks and it is safe to use? I think the OP has long left the building and gone off to buy himself some coiled wire to build himself a load tester. I hope he doesn't get hurt. I think I read that he was going to show other people how to use his method as well. Awesome. I love this forum because there are a lot of intelligent people on it, but this one has me baffled. There are tried and true method of testing batteries. Safe and sound. Why can't we just accept that and leave it alone? And just for the record, that would be one hell of a bank of 1157 bulbs to produce enough current draw to bring down a 1000 CCA battery. You could light up Central Park with that.