Two Failed Shorai Batteries While AGM Works Fine...

[jcharny]

I'm at a total loss (pun intended) with the battery situation on a 1981 CB750F build. All ignition and charging components on the bike are stock except for the regulator and rectifier (replaced with a Rick's Electrics combo unit) and the coils (replaced with Dyna 3ohm coils). The bike also has a completely custom harness with thicker gauge wire in most places, new connectors, and a 12V relay powering the coils and accesories. The harness is not yet shrink wrapped, so I could easily see if wire insulation was damaged somewhere or if there was a bad connection or ground.

1) I replaced the AGM battery with a Shorai LiFePO4 battery (LFX14A2-­BS12) about a month ago. The Shorai measured 13.3V out of the box and the bike started one or twice on the lift with the Shorai installed. After 30 minutes of working on a diffierent part of the bike, I went to start it and the battery started acting VERY strange. Symptoms were as follows:

• Voltage dropped rapidly (i.e. ~1V/15sec until my multimeter was showing <3V).
  
• If I pulled the main 30A fuse from the starter solenoid, voltage would slowly creep back up to ~10V. Voltage would plummet the second the fuse was reinserted.
  
• A battery tender could get the Shorai back to ~13V but the voltage would plummet in seconds if its was reconnected to the bike and the solenoid fuse was installed.
  
• I bypassed the started solenoid and connected the Shorai directly to the coils/accesory relay with a 10A in-line fuse. Voltage on the battery would climb to ~6V, at which point Voltage would be sufficient to trigger the relay switch. The instant the relay switch closed, Voltage would drop rapidly until the relay switch opened and Voltage would start to climb again.
  
• I hooked up an AGM battery in parallel to the Shorai to get the bike started, then disconnected the AGM and checked Voltage across the Shorai at ~5,000RPM. Voltage was ~14.5V.
  
• I disconnected every accesory circuit on the bike to check for shorts... same story, Voltage plummets to <3V if the fuse is in the starter solenoid.

2) Shorai replaced the battery on warranty with a brand new LFX14A2-­BS12. Yesterday I installed it and the sequence of events was identical. 13.3V out of the box- worked for about half an hour with steady operating and charging voltage, then the same thing started happening.

3) Completely baffled, I replaced the Shorai with an AGM battery and the bike runs great! No voltage drops. 14.4V to 14.7V across the battery terminals at 5000RPM through redline. Starts on the first push of the starter.

I'm really not sure where to go from here, given that the bike runs fine with an AGM battery. I've already contacted Shorai to get their thoughts, though I don't see why a third battery would be any different. Any insight would be appreciated.

Thanks,

JMC

 

[Kamn]

who made your custom harness??

 

[teazer]

Sounds like there is a "leak" ie a short and I wonder if it's through the Regulator/rectifier. Is the R/R connected to the starter solenoid directly? If not what else is connected to that solenoid? I would replace that and check the wiring diagram.

I'm not sure what the voltage range of a Shorai is but I think that 14.7 a touch high and I'm not sure if a battery tender is suitable for that battery. Some LiPo need a balancer and some seem to have that built in, but I'm not familiar enough with that one.

 

[jcharny]

Thank you both for the responses.

who made your custom harness??

I made the harness. All electronics are in a different location from the stock setup so it was easier to just to start from scratch.

Is the R/R connected to the starter solenoid directly? If not what else is connected to that solenoid? I would replace that and check the wiring diagram.

Besides the battery and starter motor, each connected to the appropriate solenoid post, there are also four connections on the solenoid. Two are fused via the 30A main fuse and connected to the appropriate wires coming from the regulator (Red/White and Black per the Honda wiring diagram). The other two connections on the solenoid are the trigger and the ground for the coil. If there was a short, I imagine the AGM battery would not function correctly.

I'm not sure what the voltage range of a Shorai is but I think that 14.7 a touch high and I'm not sure if a battery tender is suitable for that battery. Some LiPo need a balancer and some seem to have that built in, but I'm not familiar enough with that one.

Shorai claims that anything below 15.2V is okay and that a normal tender can be used so long as it does not have a "desulfation" mode.

 

[Sonreir]

Sounds like there is a "leak" ie a short and I wonder if it's through the Regulator/rectifier. Is the R/R connected to the starter solenoid directly? If not what else is connected to that solenoid? I would replace that and check the wiring diagram.

I'm not sure what the voltage range of a Shorai is but I think that 14.7 a touch high and I'm not sure if a battery tender is suitable for that battery. Some LiPo need a balancer and some seem to have that built in, but I'm not familiar enough with that one.

+1 on both points.

With the key turned off and your AGM battery installed, disconnect the negative terminal of the battery and use a multimeter to measure amperage between the removed ground cable and the negative terminal of the battery. If your meter doesn't have an amp setting, use voltage.

Let us know the number you got from the reading.

 

[jcharny]

+1 on both points.

With the key turned off and your AGM battery installed, disconnect the negative terminal of the battery and use a multimeter to measure amperage between the removed ground cable and the negative terminal of the battery. If your meter doesn't have an amp setting, use voltage.

Let us know the number you got from the reading.

I performed the test and I have a parasitic draw of 2.5A coming from my regulator. I can stop the draw if I remove the black wire from the regulator. NOTE: the black wire is also the only wire that I have connected differently from the stock diagram. Below are diagrams of the regulator circuit in the stock and modified configurations. I do not have a key on my bike, just a mini toggle switch which triggers a relay and provides power to my accesories and coils. As I don't have a key, I figured I would just run the black wire directly back to the solenoid.

What am I missing here? How does this circuit (stock or modified) even work if it starts and ends at the positive terminal of the battery? My understanding of voltage regulators is that there is usually three connections, an input for unregulated voltage, an output for regulated voltage, and a ground. If the input and output are both connected to the same voltage source, how is the voltage output from the alternator even entering the circuit?

 

[Sonreir]

You're not taking into account the black and white wires that run from the R/R to the alternator. When the black wire from the R/R has power, it sends that power through the other black wire, to the alternator. Your alternator is an electromagnet type, not a permanent magnet type, so it needs power to create power. The regulator on your bike works by adjusting the resistance on the white wire coming back into the unit. If you bike isn't producing enough power, the resistance is lowered. This allows more power to pass through the electromagnet and create a strong magnetic field, in turn creating more electricity.

However, this only works if your engine is spinning. With everything hooked up and the engine at a stand still, no power is being produced. With no power being produced, the regulator sends as much juice as it can to your alternator in the hopes that it can make up the difference.

Short version: Don't hook it up the way you have it hooked up. Use the stock setup.

 

[jcharny]

However, this only works if your engine is spinning. With everything hooked up and the engine at a stand still, no power is being produced. With no power being produced, the regulator sends as much juice as it can to your alternator in the hopes that it can make up the difference.

Thanks Sonreir! I'll change the wiring so the black wire to the regulator only gets power when my ignition switch is turned on, but I still think I'm going to have a problem. The 2.5A drain will just show up when I turn the bike on (I actually confirmed this yesterday by switching the wiring back to stock). If this is what's causing my lithium battery to rapidly discharge, I would need to start the bike within seconds of turning on the ignition switch. That is, wiring it in the stock method may fix the parasitic draw when my bike is off, preventing the battery from discharging when the bike is sitting, but I don't think it will solve the larger battery problem I'm having.

 

[Sonreir]

A 2.5A draw should take around three to four hours to drain your battery, assuming you're using the correctly sized battery. I wouldn't worry about having to start the bike too quickly unless you're facing some other issues as well.

 

[grcamna5]

subscribed

 

[jcharny]

A 2.5A draw should take around three to four hours to drain your battery, assuming you're using the correctly sized battery.

I just got off the phone with Shorai's head tech. The LFX14A2-BS12 has the equivalent CCA (210) of a traditional 12Ah lead-acid battery. However, the Shorai ony has a capacity of approximately 4Ah. I think we've found the problem... if there was a parasitic draw of 2.5A when the bike when the ignition switch was not on, the battery would be completely discharged in under 2 hours. In my first post, I mentioned that everything was fine for the first 30 minutes after the Shorai was installed. The battery was sufficiently discharged to potentially damage the lithium cells, now requiring that they be rebalanced.

I'm not yet sure that the case is closed on this one, but definitely making progress towards a solution. Moral of the story thus far is that a parisitic draw will be much more noticable on a lithium battery, given its lower capacity, and potentially damaging. Once a lithium battery is discharged below 12V, things get dicey.

 

[Sonreir]

I just got off the phone with Shorai's head tech. The LFX14A2-BS12 has the equivalent CCA (210) of a traditional 12Ah lead-acid battery. However, the Shorai ony has a capacity of approximately 4Ah. I think we've found the problem... if there was a parasitic draw of 2.5A when the bike when the ignition switch was not on, the battery would be completely discharged in under 2 hours. In my first post, I mentioned that everything was fine for the first 30 minutes after the Shorai was installed. The battery was sufficiently discharged to potentially damage the lithium cells, now requiring that they be rebalanced.

I'm not yet sure that the case is closed on this one, but definitely making progress towards a solution. Moral of the story thus far is that a parisitic draw will be much more noticable on a lithium battery, given its lower capacity, and potentially damaging. Once a lithium battery is discharged below 12V, things get dicey.

Actually... they can get as low as 9V without any serious issues, but best to keep 'em closer to 12V as much as is possible.

 

[jcharny]

You're not taking into account the black and white wires that run from the R/R to the alternator. When the black wire from the R/R has power, it sends that power through the other black wire, to the alternator.

Assuming the wiring harness is stock, should I see 12.5V on the black wire coming from the regulator when the ignition switch is on but the bike is not running?

 

[Sonreir]

Assuming the wiring harness is stock, should I see 12.5V on the black wire coming from the regulator when the ignition switch is on but the bike is not running?

You'll always see the reference voltage on the black wire if the ignition switch is on. The engine on or off won't make a difference. The reference voltage can and will change, though. When the engine is not running, the reference voltage is whatever the battery voltage is. When the engine is running, reference voltage will be whatever the alternator is putting out (so long as that amount of electricity isn't being fully consumed by your motorcycle's electrical system). For instance, when you activate the starter motor, it's not uncommon to see the reference voltage drop to 10V or even 9V, but it will jump right back up into the 12s once you stop running the starter motor.

The white and black wires running to/from the alternator are their own circuit. The black wire will always have full voltage, but how much makes it through the circuit is determined by the resistance on the white wire.

If you think of it in terms of plumbing, the black wire is the water main. It has whatever pressure (voltage) it has. How much total water (electricity) makes it through your house is determined by the valve you have open or closed (this is the internal functionality of the r/r on the white wire). The "valve" inside the regulator is what determines how much power your bike produces. It does this by sensing the reference voltage on the black wire (which changes based on how much power is being generated versus how much power is being used) and then either opening the "valve" (reducing the resistance in the circuit) to allow more power through the exciter coil (this creates a strong magnetic field and so your alternator produces more electricity at the same engine speed) or closing the valve (increasing the resistance in the circuit) to allow less power through the exciter coil (this reduces the strength of the magnet field, meaning less power is produced by the alternator).

 

[jcharny]

Thanks for the explanation Sonreir. For whatever reason, the reg/rect combo is still super confusing to me.

I currently have a normally-closed SPST relay with two outputs (87a and 87b). 87a powers all my accesories and 87b powers my coils. Instead of putting a new relay on the bike, could I connect the black wire from the regulator to one of those two outputs via a splice?

 

[Sonreir]

I'd double check that relay. SPST relays only have one output. DPST relays have two outputs. SPDT relays have two outputs, but only one is on at a time (and one is always on, there is no off position for both).

 

[jcharny]

It has two ouputs- diagram is below (what I am calling 87a is labeled simply as 87).



What do you think Sonreir, will a splice work?

 

[Sonreir]

So 87b is going to the coils and 87 is everything else?

Also, it's academic, but 87a is in reference to a SPDT relay, which has 87 and 87a. SPST relays have only 87.

 

[jcharny]

So 87b is going to the coils and 87 is everything else?

Correct- 87b is strictly the coils and 87 is a single wire going to a fuse block in my headlight bucket. Could I splice the black wire from the regulator to the output wire from 87?

 

[Sonreir]

Yup. That's what I would do.