Why is there zero power after trying to start with a loose battery connection?

I know this is pretty much always the way it is, so the behavior doesn't surprise me, but I'm trying to troubleshoot an electrical issue that I believe might be a bad battery cable and I think I need to understand more about how stuff works to be able to think this through.

Here's what's up:  On the motorhome, sometimes when you turn the key to "run" the dash lights come on, the voltmeter reads 12.2 volts and all the chassis-connected devices work fine.  As soon as you turn the key to crank, nothing happens under the doghouse but everything else goes dark.  It all stays dark until you wiggle the battery cable and it all comes back on.  Rinse/repeat until you manage to get it into that magical position where it will actually start.  I get why the loose connection won't allow the starter to turn, but why is it that when you do try to crank it suddenly the other things that were working stop?

The other weird thing that is happening is that with the motor running, I try to operate the leveling jacks which use a hydraulic pump which is driven off of the chassis 12v system.  As soon as I try to deploy the jacks it will kill the engine and all the chassis 12v things go dark until wiggling the cable.  Part one of my plan was to replace the battery cable and clean up the ground connection to the frame, but I'm not sure if that's the total answer.  I feel like I need to understand why everything just goes dead so I can think through the rest of it. 

Make sure all of our cable connections are clean and tight. Then redo your tests.I would bet that things work better then.

To me it sounds like a bad cell in the battery.  It can't handle the load.

How old is the cable. They do go bad. Corrosion inside the cable. I would replace both the cable to the starter solenoid and the ground. If there is a cable from the solenoid to the starter replace that as well. 

The reason it goes dead is because in the start position power is only directed to the ignition and the starter. If it is a electric pump that gets power as well if it is EFI then the computer gets power as well but all accessories get cut so All power goes to the starter. 

It really sounds like you have a bad cable and not just a bad connection.

The only other thing I can think of is a bad solenoid Alf lastly a dead spot on the starter but with the leveling hydro Lucas also triggering it I would start with the cables.  

Do you have a spare battery that is good to try? It might not solve the problem, but you could then rule out a wonky battery.

They stop because the load from the starter causes an arc at the loose battery terminal.

Once it arcs, the connection is burned and won't pass current.

Wiggling it rubs the two surfaces together and lets it make contact again, providing enough current to run low-load things like dash lights but not enough for high-load things like the starter.

When you hit the starter again, the whole process starts over.

Is there some reason that you're not tightening the connection?

Rules for electrical are clean, dry, tight.

So far, I've already gone through the connection-tightening process which is why I'm replacing the cable.  

So, if it's an issue of things arcing then that starts to make more sense.   I assume the hydraulic pump has a pretty heavy draw, but I need to follow its grounding to see how it's managing to kill the engine.

If your voltage meter is accurate, then 12.2 V is a dead battery. A strong battery should read over 13V to mid 14V. 

I had a similar issue with my tractor and 30+ year old connectors. It acted like a dead battery, so I decided to try to clean the battery connections first. I broke one in the process, so I bolted on two new ones. After that it started like it had a brand new battery. 

A poor connection may also explain the low voltage. You could confirm the reading with a voltage meter. 

Trans maro has the story right as to why you lose power.

Check the voltage at the battery terminal and then do the same at the terminal. Chances are the battery stays at 12 volts while the power is disappearing under load at the connector. 

Go find a video on "voltage drop testing" and watch it. 

Don't just tighten the connections, take them apart and clean them.

I chased a problem like this on my old pickup truck.  It stopped when I changed the battery cables. 

Not saying this is your problem but it sure sounds similar to mine.

I get into this fairly often on the equipment i work on. Had it on my boat too. The weak connection works for light stuff since the amp load is small. The starter pulls far more amps than the ignition and indicator lights.  The voltage drops, how far depends on resistance and amp load. Testing can be done with jumper cable. Use them to bypass current connections and see if the problem goes away. Go from batter terminal to the source. Test circuit by circuit if you can. When i magically works, you found your bad circuit. It could be common at the battery or just the starter.

On my boat the wiring was probably double the length it needed, connections poor. It would crank and crank, kinda hit then die, maybe start ok and keep running fine all day. I used a meter and watched the input voltage to the computer(efi motor, suzuki df50), the crummy connections and other resistance causing problems had my 13.4 battery voltage dropping to under 6v at the computer when the starter engaged. Computers dont get really happy when you feed them half the power and the half it gets is dirty. 

I had added another large good battery in parallel with the boat battery and it made no real difference. The wiring was that bad. Tossed the extra wiring(they actually had a factory extension in line that was totally unnecessary). More length, more connections/more resistance, more amps= less voltage. It still has a minor issue with it but nothing like it was. 

Yeah, I've gone in to this with the assumption that the likely problem is a bad battery cable - I've had that problem on a couple different cars and 25 years is a pretty long time for something that sits outside all the time and gets bounced around to not get any corrosion or breakage.  Plus, the first thing I did was to check the tightness of the connections to the battery and to the frame.  The frame connection has some rust around it, but the connection was solid so I left it alone -- I was 740 miles from home and didn't have my flapwheel.  The battery connection was a little loose and by giving it about a turn and a half or so, it was tight and the problem with starting and with deploying the automatic steps went away completely.  Three starts and 740 miles later, everything went to heck again when I tried to deploy the jacks and it killed the motor.  I'm checking out what seems to be a pretty good video on voltage drop theory and testing now and I think that's going to help a bit.  The whole thing is also complicated by the fact that the ground is shared with the house battery systems and those cables don't seem so great either, so I'm going to remove the house batteries from the equation and try some measurements and tests.

In reply to Nate90LX :

Wrong.  12.2/12.5 is a fully charged battery.. 14.3 is the designed charging rate of the alternator.

The instruments etc. do not draw much amperage so a bad cable would work but once the heavy draw of the starter the cable cannot handle the load.

I've cut old suspect cables open to find the green schmoo of death. It got to be routine to just replace any older cables on a new to me vehicle.

Clean all connections, each and every single one, not just the mains. 

This got to be routine w/ me.

iceracer said:

In reply to Nate90LX :

Wrong.  12.2/12.5 is a fully charged battery.. 14.3 is the designed charging rate of the alternator.

I'd say you're both wrong.  A fully charged 12v lead acid battery that's had no loads on it and has had time to stabilize after charging should be anywhere from 12.6 - 13 volts depending on the battery (some AGMs hold higher resting voltages), assuming it's around 70 - 80* outside.  Cold weather will show lower voltage, but 12.2 volts in warm weather is only about half charged (assuming it's been resting for a bit). 

If you're not familiar with voltage drop testing, i'd study up on that so you can actually test some of these connections 'under load'.  All you need to do it really is a meter with a 'min/max' or 'peak recall' function. 

Trans_Maro said:

They stop because the load from the starter causes an arc at the loose battery terminal.

Once it arcs, the connection is burned and won't pass current.

Wiggling it rubs the two surfaces together and lets it make contact again, providing enough current to run low-load things like dash lights but not enough for high-load things like the starter.

When you hit the starter again, the whole process starts over.

Is there some reason that you're not tightening the connection?

Rules for electrical are clean, dry, tight.

I had this exact problem on my truck, terminals seemed tight but after a few tries and really weird electrical issues happening i eventually found the ground cable to be warm to the touch which lead me to look at the terminal more closely and noticed it was cracked, new terminal fixed it right away

In reply to rslifkin :

Many years experience cause me to not agree.