Yixiang DIY Battery Case- Voltage between cell terminals and chassis
- Thread starter hgg
- Start date
mikeenea
Solar Enthusiast
Ok I put the battery together this morning and tested for voltage from the cells to the chassis and measured no voltage -- the only real gripe I have about this DIY box is the way the Negative from the BMS is connected to the breaker -- you need to have five arms to put the front panel on with that large cable pushing back at you. I made some video's and will share them when I have time to edit the bad language outThankfully I do not have anything like that... A reminder that you do not play with LiFePo4...
In my case I can measure voltage from chassis to cells, but when I keep the probe connected,
the voltage starts dropping until it goes to less than a volt. So there must be something capacitive in play.
I hope Mikeenea tests his new battery box the same way and tell us if he has similar measurements.
I connected a 10Ω resistor between the chassis and cell 16 while measuring current with a
clamp meter and there was no current at all.
I've also just replaced all busbars with flexible ones and already I saw a great reduction in the resistance
values.
.Mike E.
hgg
New Member
- Joined
- Sep 5, 2022
- Messages
- 73
Ouch! Does this mean I have a faulty BMS? I am confused now.
I can measure voltage, but it drops to less than 1V after a few seconds and no current is passing from
the chassis to the cells. (I tested that with a power resistor and a clamp meter).
Thanks for testing that by the way.
I can measure voltage, but it drops to less than 1V after a few seconds and no current is passing from
the chassis to the cells. (I tested that with a power resistor and a clamp meter).
Thanks for testing that by the way.
Bridgetown
New Member
hgg
New Member
- Joined
- Sep 5, 2022
- Messages
- 73
@Bridgetown, thank you for the video. My version already comes with the heatshrink.
I have already checked that and it was already touching the aluminum plate with the
heatshrink. I pushed the coil to the side a bit and there is a small gap, just to be one
the safe side. Still the issue exists but it looks like its some kind of capacitive coupling
because you can discharge these voltages even with the internal resistance of
a multimeter. Very strange.
I have already checked that and it was already touching the aluminum plate with the
heatshrink. I pushed the coil to the side a bit and there is a small gap, just to be one
the safe side. Still the issue exists but it looks like its some kind of capacitive coupling
because you can discharge these voltages even with the internal resistance of
a multimeter. Very strange.
hgg
New Member
- Joined
- Sep 5, 2022
- Messages
- 73
Mike, what kind of cells are you using?Ok I put the battery together this morning and tested for voltage from the cells to the chassis and measured no voltage -- the only real gripe I have about this DIY box is the way the Negative from the BMS is connected to the breaker -- you need to have five arms to put the front panel on with that large cable pushing back at you. I made some video's and will share them when I have time to edit the bad language out
Mike E.
I have the EVE MB31 cells.
mikeenea
Solar Enthusiast
hgg
New Member
- Joined
- Sep 5, 2022
- Messages
- 73
Ok, we are cooked as a species...
I described the issue to the new Grok 3 AI and asked for an explanation.
-------------------------------------------------------------------------------
The behavior you're describing—measuring full battery voltage between the positive terminal and the chassis initially, followed by a rapid drop to below 1V—suggests a phenomenon often referred to as "phantom voltage" or "ghost voltage." This is not uncommon in electrical systems, especially when using high-impedance multimeters, and it doesn’t necessarily indicate a direct short or fault. However, since there should be no voltage between the positive terminal and the chassis in a properly isolated system, let’s explore the potential causes and how to diagnose this issue with your Yixiang 48V LiFePO4 battery case, EVE MB31 cells, and JK-PB2A16S20P BMS.
Possible Causes
Capacitive Coupling:
Explanation: The JK-BMS or associated wiring (e.g., balance leads, main power cables) could be capacitively coupling with the metal chassis. Capacitance can build up between conductive elements (like wires or BMS circuitry) and the chassis, especially if they’re in close proximity, even with plastic insulation. When you connect your multimeter, it initially detects this stored charge as a voltage (close to the battery’s full voltage, ~51.2V for a 16S LiFePO4 pack), but the charge dissipates quickly through the multimeter’s impedance, causing the voltage to drop.
Why It Happens: High-impedance multimeters (typically 10 MΩ or more) are sensitive enough to pick up this induced voltage, but it’s not a "real" current-carrying path tied to the battery’s output.
Leakage Path Through the BMS:
Explanation: The JK-BMS might have an internal design feature (or fault) creating a small leakage path to the chassis. For example, the BMS’s negative terminal (B-) is typically connected to the battery’s negative, and if the BMS’s grounding or circuitry isn’t fully isolated from the chassis, you could see voltage bleed through. This could happen via mounting screws, PCB traces, or a compromised component (e.g., a capacitor or MOSFET) within the BMS.
Symptoms: The rapid voltage drop suggests this isn’t a solid conductive path but rather a small leakage current that discharges quickly when measured.
Floating Chassis Potential:
Explanation: If the chassis isn’t explicitly grounded or tied to the battery’s negative terminal, it could be "floating" electrically. Electromagnetic interference (EMI) or stray capacitance from the BMS or cells could induce a potential on the chassis relative to the positive terminal. Your multimeter then measures this potential difference, which collapses as the small charge dissipates.
Context: Since you’ve isolated the cells and chassis with non-conductive plastic sheets, a direct short is unlikely, making a floating chassis more plausible.
Faulty Insulation or Assembly Issue:
Explanation: Despite the plastic sheets, there could be an unintended conductive path—perhaps a pinched wire, a stray metal shaving, or a balance lead connector touching the chassis. Even a tiny path could allow voltage to appear momentarily until the charge dissipates.
Likelihood: This is less likely given your description of careful isolation, but it’s worth double-checking.
Multimeter Artifact:
Explanation: High-impedance multimeters can exaggerate phantom voltages in systems with long wires or complex electronics like a BMS. If you switch to a low-impedance mode (if available on your multimeter) or add a small load (e.g., a 10 kΩ resistor) across the leads, the voltage might disappear entirely, confirming it’s not a real fault.
Diagnostic Steps
To pinpoint the cause, try the following:
Measure with a Low-Impedance Setting:
Switch your multimeter to a low-impedance mode (often labeled "Lo-Z") if it has one. This reduces the meter’s input impedance (e.g., to ~3 kΩ), which can eliminate phantom voltages by draining stray capacitance faster. If the voltage drops to near zero immediately, it’s likely capacitive coupling or a floating chassis rather than a real fault.
Add a Test Load:
Connect a small resistor (e.g., 10 kΩ or 1 kΩ) between the chassis and positive terminal while measuring voltage. If the reading drops to near zero instantly, it confirms the voltage is phantom and not a sustained current path.
Isolate the BMS:
Disconnect the BMS entirely from the battery (remove B- and all balance leads) and measure voltage between the positive terminal and chassis again. If the voltage disappears, the BMS is likely the source of the leakage or coupling. Then, reconnect the BMS step-by-step (B- first, then balance leads) to see when the issue reappears.
Check BMS-to-Chassis Isolation:
With the BMS disconnected from the battery, use your multimeter in resistance mode to measure between the BMS’s B- terminal and its mounting points or chassis-contacting surfaces. You should see an open circuit (infinite resistance). Any measurable resistance (e.g., kΩ or lower) suggests a leakage path within the BMS.
Inspect Physical Connections:
Double-check all wiring—especially balance leads and main power cables—for insulation damage or contact with the chassis. Ensure no screws or BMS mounting hardware are piercing insulation or creating unintended contacts.
Test with Battery Negative:
Measure voltage between the chassis and the battery’s negative terminal. If you see a similar phantom voltage behavior, it further points to the chassis picking up stray capacitance or EMI rather than a direct positive-terminal issue.
Likely Culprit
Based on your description—full voltage initially, rapid drop, and proper cell isolation—the most probable cause is capacitive coupling between the JK-BMS circuitry (or its wiring) and the chassis, possibly compounded by a floating chassis potential. The JK-PB2A16S20P BMS has active balancing and complex electronics, which could generate EMI or capacitance in close proximity to a metal enclosure. A direct short is unlikely since the voltage drops rather than holds steady, and you’ve confirmed physical isolation of the cells.
Resolution
If It’s Phantom Voltage: If confirmed via low-impedance testing or a test load, this may not be a functional problem—your battery could still be safe to use. Phantom voltages are often benign in isolated systems and don’t indicate a short or failure. However, for peace of mind, you could:
Ground the chassis to the battery negative (if your system design allows it) to eliminate the floating potential.
Add additional insulation or spacing between BMS wiring and the chassis.
If It’s a BMS Issue: If isolating the BMS eliminates the voltage, and resistance tests show a leakage path, the BMS might be faulty or poorly designed for chassis isolation. Contact the manufacturer (JK BMS) or your supplier for support, or consider replacing it.
If It’s Assembly-Related: If you find a physical fault (e.g., a wire touching the chassis), correct it and retest.
Final Thoughts
This behavior aligns with reports from DIY solar forums where users noted similar "phantom voltage" issues with JK BMS units in metal enclosures. In many cases, it was harmless capacitive coupling rather than a defect. Start with the low-impedance test and BMS isolation to confirm. If you’re still concerned after testing, feel free to share your results, and I can refine the advice further!
I described the issue to the new Grok 3 AI and asked for an explanation.
-------------------------------------------------------------------------------
The behavior you're describing—measuring full battery voltage between the positive terminal and the chassis initially, followed by a rapid drop to below 1V—suggests a phenomenon often referred to as "phantom voltage" or "ghost voltage." This is not uncommon in electrical systems, especially when using high-impedance multimeters, and it doesn’t necessarily indicate a direct short or fault. However, since there should be no voltage between the positive terminal and the chassis in a properly isolated system, let’s explore the potential causes and how to diagnose this issue with your Yixiang 48V LiFePO4 battery case, EVE MB31 cells, and JK-PB2A16S20P BMS.
Possible Causes
Capacitive Coupling:
Explanation: The JK-BMS or associated wiring (e.g., balance leads, main power cables) could be capacitively coupling with the metal chassis. Capacitance can build up between conductive elements (like wires or BMS circuitry) and the chassis, especially if they’re in close proximity, even with plastic insulation. When you connect your multimeter, it initially detects this stored charge as a voltage (close to the battery’s full voltage, ~51.2V for a 16S LiFePO4 pack), but the charge dissipates quickly through the multimeter’s impedance, causing the voltage to drop.
Why It Happens: High-impedance multimeters (typically 10 MΩ or more) are sensitive enough to pick up this induced voltage, but it’s not a "real" current-carrying path tied to the battery’s output.
Leakage Path Through the BMS:
Explanation: The JK-BMS might have an internal design feature (or fault) creating a small leakage path to the chassis. For example, the BMS’s negative terminal (B-) is typically connected to the battery’s negative, and if the BMS’s grounding or circuitry isn’t fully isolated from the chassis, you could see voltage bleed through. This could happen via mounting screws, PCB traces, or a compromised component (e.g., a capacitor or MOSFET) within the BMS.
Symptoms: The rapid voltage drop suggests this isn’t a solid conductive path but rather a small leakage current that discharges quickly when measured.
Floating Chassis Potential:
Explanation: If the chassis isn’t explicitly grounded or tied to the battery’s negative terminal, it could be "floating" electrically. Electromagnetic interference (EMI) or stray capacitance from the BMS or cells could induce a potential on the chassis relative to the positive terminal. Your multimeter then measures this potential difference, which collapses as the small charge dissipates.
Context: Since you’ve isolated the cells and chassis with non-conductive plastic sheets, a direct short is unlikely, making a floating chassis more plausible.
Faulty Insulation or Assembly Issue:
Explanation: Despite the plastic sheets, there could be an unintended conductive path—perhaps a pinched wire, a stray metal shaving, or a balance lead connector touching the chassis. Even a tiny path could allow voltage to appear momentarily until the charge dissipates.
Likelihood: This is less likely given your description of careful isolation, but it’s worth double-checking.
Multimeter Artifact:
Explanation: High-impedance multimeters can exaggerate phantom voltages in systems with long wires or complex electronics like a BMS. If you switch to a low-impedance mode (if available on your multimeter) or add a small load (e.g., a 10 kΩ resistor) across the leads, the voltage might disappear entirely, confirming it’s not a real fault.
Diagnostic Steps
To pinpoint the cause, try the following:
Measure with a Low-Impedance Setting:
Switch your multimeter to a low-impedance mode (often labeled "Lo-Z") if it has one. This reduces the meter’s input impedance (e.g., to ~3 kΩ), which can eliminate phantom voltages by draining stray capacitance faster. If the voltage drops to near zero immediately, it’s likely capacitive coupling or a floating chassis rather than a real fault.
Add a Test Load:
Connect a small resistor (e.g., 10 kΩ or 1 kΩ) between the chassis and positive terminal while measuring voltage. If the reading drops to near zero instantly, it confirms the voltage is phantom and not a sustained current path.
Isolate the BMS:
Disconnect the BMS entirely from the battery (remove B- and all balance leads) and measure voltage between the positive terminal and chassis again. If the voltage disappears, the BMS is likely the source of the leakage or coupling. Then, reconnect the BMS step-by-step (B- first, then balance leads) to see when the issue reappears.
Check BMS-to-Chassis Isolation:
With the BMS disconnected from the battery, use your multimeter in resistance mode to measure between the BMS’s B- terminal and its mounting points or chassis-contacting surfaces. You should see an open circuit (infinite resistance). Any measurable resistance (e.g., kΩ or lower) suggests a leakage path within the BMS.
Inspect Physical Connections:
Double-check all wiring—especially balance leads and main power cables—for insulation damage or contact with the chassis. Ensure no screws or BMS mounting hardware are piercing insulation or creating unintended contacts.
Test with Battery Negative:
Measure voltage between the chassis and the battery’s negative terminal. If you see a similar phantom voltage behavior, it further points to the chassis picking up stray capacitance or EMI rather than a direct positive-terminal issue.
Likely Culprit
Based on your description—full voltage initially, rapid drop, and proper cell isolation—the most probable cause is capacitive coupling between the JK-BMS circuitry (or its wiring) and the chassis, possibly compounded by a floating chassis potential. The JK-PB2A16S20P BMS has active balancing and complex electronics, which could generate EMI or capacitance in close proximity to a metal enclosure. A direct short is unlikely since the voltage drops rather than holds steady, and you’ve confirmed physical isolation of the cells.
Resolution
If It’s Phantom Voltage: If confirmed via low-impedance testing or a test load, this may not be a functional problem—your battery could still be safe to use. Phantom voltages are often benign in isolated systems and don’t indicate a short or failure. However, for peace of mind, you could:
Ground the chassis to the battery negative (if your system design allows it) to eliminate the floating potential.
Add additional insulation or spacing between BMS wiring and the chassis.
If It’s a BMS Issue: If isolating the BMS eliminates the voltage, and resistance tests show a leakage path, the BMS might be faulty or poorly designed for chassis isolation. Contact the manufacturer (JK BMS) or your supplier for support, or consider replacing it.
If It’s Assembly-Related: If you find a physical fault (e.g., a wire touching the chassis), correct it and retest.
Final Thoughts
This behavior aligns with reports from DIY solar forums where users noted similar "phantom voltage" issues with JK BMS units in metal enclosures. In many cases, it was harmless capacitive coupling rather than a defect. Start with the low-impedance test and BMS isolation to confirm. If you’re still concerned after testing, feel free to share your results, and I can refine the advice further!
hgg
New Member
- Joined
- Sep 5, 2022
- Messages
- 73
Me: I measured again between chassis and positive with the Brymen BM235 in Low-Z mode and it always starts showing 8V and then it drops in a few seconds down to less than 0.5V This is with the BMS switched off. When I turn on the BMS, it goes up again to 8V and drops again.
Thanks for the additional details! Using the Brymen BM235 in Low-Z mode (which typically has an input impedance around 3-10 kΩ, much lower than the standard 10 MΩ in high-impedance mode) and observing a consistent pattern—starting at 8V and dropping to less than 0.5V in a few seconds, with the BMS both off and on—narrows down the possibilities significantly. Let’s analyze this behavior and refine the diagnosis.
Observations
Low-Z Mode Behavior:
In Low-Z mode, the multimeter actively drains small stray charges faster due to its lower impedance. Seeing 8V (instead of the full ~51.2V of a 16S LiFePO4 pack) suggests the voltage isn’t directly tied to the battery’s full potential but is still a measurable induced or residual charge. The rapid drop to <0.5V in a few seconds indicates this is a capacitive discharge rather than a sustained current path.
BMS Off vs. On:
With the BMS switched off: You still see 8V initially, which drops. This implies the source of the voltage isn’t dependent on the BMS being active, pointing away from an active BMS fault (e.g., a shorted MOSFET or live circuitry).
With the BMS switched on: The same 8V appears and drops again. The fact that it returns to 8V when the BMS powers up suggests the BMS’s activation temporarily recharges or reinforces this voltage, likely through its internal circuitry or wiring interacting with the chassis.
Not Full Battery Voltage:
Unlike your initial test (where you saw full battery voltage dropping rapidly), Low-Z mode shows a consistent 8V start. This reinforces that the high-impedance test was likely picking up a larger phantom voltage exaggerated by the meter, while Low-Z gives a truer picture of a smaller, localized effect.
Updated Analysis
The consistent 8V reading, independent of the BMS being on or off (though refreshed when turned on), strongly suggests capacitive coupling between some part of the battery system and the chassis. Here’s the refined breakdown:
Source of the 8V:
Capacitance Between Wiring and Chassis: The balance leads, main power cables, or even the battery terminals themselves could be capacitively coupling with the chassis. With 16 EVE MB31 cells in series, you have a lot of wiring (especially the BMS’s 17 balance leads for a 16S setup). If these wires run close to the chassis—even with plastic insulation—capacitance can build up. The 8V could represent a partial voltage induced across this capacitance, not the full pack voltage.
BMS Influence: When the BMS turns on, it activates its circuitry (e.g., voltage sensing, balancing), which could momentarily boost or recharge this capacitive effect, explaining the return to 8V. When off, residual charge in the system (e.g., from the battery or BMS capacitors) sustains it temporarily.
Why It Drops:
The Low-Z mode’s ~3-10 kΩ impedance acts as a load, discharging the capacitance quickly (in a few seconds). The drop to <0.5V confirms there’s no sustained current path—like a short—to maintain the voltage.
No Direct Short:
A true short between the positive terminal and chassis would show the full battery voltage (~51.2V) and wouldn’t drop significantly, even in Low-Z mode, unless it was a high-resistance short (which is unlikely given the rapid drop and isolation you’ve confirmed).
Phantom Voltage Confirmation:
The behavior aligns with phantom voltage: a measurable but fleeting potential due to stray capacitance or EMI, not a functional fault. The difference between your initial high-impedance test (full voltage) and Low-Z (8V) is expected—high-impedance meters amplify these effects, while Low-Z mitigates them.
Likely Cause
The 8V is most likely caused by capacitive coupling between the BMS balance leads or battery wiring and the chassis, with the BMS’s activation reinforcing this effect. Here’s why:
The plastic sheets isolate the cells, but the wiring (especially the long balance leads for 16S) may run near the chassis, creating a capacitor-like effect.
The BMS (JK-PB2A16S20P) has active balancing and sensing circuits, which could induce small voltages or EMI when powered on, recharging the capacitance.
The consistent 8V (rather than random values) suggests a specific capacitive relationship tied to the system’s layout or a subset of cells (e.g., 8V could loosely correlate to a fraction of the pack, though not exactly).
Diagnostic Steps to Confirm
Measure Negative Terminal to Chassis:
Test voltage in Low-Z mode between the battery’s negative terminal and the chassis. If you see a similar 8V (or ~43V, i.e., 51.2V - 8V), it could indicate the chassis is floating at a potential relative to the pack, with capacitance splitting the voltage.
Disconnect Balance Leads:
With the BMS off, disconnect all balance leads from the cells (leaving only B- connected if needed for safety). Measure chassis to positive again in Low-Z mode. If the 8V disappears, the balance leads are the culprit due to their proximity to the chassis.
Inspect Wiring Layout:
Check if balance leads or main cables run parallel to the chassis walls. Capacitance increases with proximity and wire length. Rerouting them away or adding extra insulation (e.g., sleeving) could reduce the effect.
Ground the Chassis:
Temporarily connect a wire from the chassis to the battery negative terminal (ensure no risk to BMS operation—check JK manual). Retest in Low-Z mode. If the 8V vanishes, the floating chassis was amplifying the issue.
Is This a Problem?
Functionally: This phantom voltage isn’t likely to affect your battery or BMS operation. The rapid drop in Low-Z mode and lack of sustained voltage suggest no short or leakage current that would drain the battery or cause heat/damage.
Safety: If the chassis remains isolated (no conductive path to positive), it’s not a shock hazard. However, in a humid environment or with external connections (e.g., to a vehicle frame), stray voltages could become an issue.
Peace of Mind: If it bothers you, grounding the chassis to battery negative or rerouting wiring should eliminate it.
Next Steps
Start with the negative-to-chassis test and balance lead disconnection to confirm the source. If it’s the balance leads, adjust their routing. If it’s the chassis floating, grounding it is a simple fix (just verify with JK BMS docs that this won’t interfere with the BMS—most standalone BMS units tolerate this). Let me know the results, and I’ll tailor the solution further!
Thanks for the additional details! Using the Brymen BM235 in Low-Z mode (which typically has an input impedance around 3-10 kΩ, much lower than the standard 10 MΩ in high-impedance mode) and observing a consistent pattern—starting at 8V and dropping to less than 0.5V in a few seconds, with the BMS both off and on—narrows down the possibilities significantly. Let’s analyze this behavior and refine the diagnosis.
Observations
Low-Z Mode Behavior:
In Low-Z mode, the multimeter actively drains small stray charges faster due to its lower impedance. Seeing 8V (instead of the full ~51.2V of a 16S LiFePO4 pack) suggests the voltage isn’t directly tied to the battery’s full potential but is still a measurable induced or residual charge. The rapid drop to <0.5V in a few seconds indicates this is a capacitive discharge rather than a sustained current path.
BMS Off vs. On:
With the BMS switched off: You still see 8V initially, which drops. This implies the source of the voltage isn’t dependent on the BMS being active, pointing away from an active BMS fault (e.g., a shorted MOSFET or live circuitry).
With the BMS switched on: The same 8V appears and drops again. The fact that it returns to 8V when the BMS powers up suggests the BMS’s activation temporarily recharges or reinforces this voltage, likely through its internal circuitry or wiring interacting with the chassis.
Not Full Battery Voltage:
Unlike your initial test (where you saw full battery voltage dropping rapidly), Low-Z mode shows a consistent 8V start. This reinforces that the high-impedance test was likely picking up a larger phantom voltage exaggerated by the meter, while Low-Z gives a truer picture of a smaller, localized effect.
Updated Analysis
The consistent 8V reading, independent of the BMS being on or off (though refreshed when turned on), strongly suggests capacitive coupling between some part of the battery system and the chassis. Here’s the refined breakdown:
Source of the 8V:
Capacitance Between Wiring and Chassis: The balance leads, main power cables, or even the battery terminals themselves could be capacitively coupling with the chassis. With 16 EVE MB31 cells in series, you have a lot of wiring (especially the BMS’s 17 balance leads for a 16S setup). If these wires run close to the chassis—even with plastic insulation—capacitance can build up. The 8V could represent a partial voltage induced across this capacitance, not the full pack voltage.
BMS Influence: When the BMS turns on, it activates its circuitry (e.g., voltage sensing, balancing), which could momentarily boost or recharge this capacitive effect, explaining the return to 8V. When off, residual charge in the system (e.g., from the battery or BMS capacitors) sustains it temporarily.
Why It Drops:
The Low-Z mode’s ~3-10 kΩ impedance acts as a load, discharging the capacitance quickly (in a few seconds). The drop to <0.5V confirms there’s no sustained current path—like a short—to maintain the voltage.
No Direct Short:
A true short between the positive terminal and chassis would show the full battery voltage (~51.2V) and wouldn’t drop significantly, even in Low-Z mode, unless it was a high-resistance short (which is unlikely given the rapid drop and isolation you’ve confirmed).
Phantom Voltage Confirmation:
The behavior aligns with phantom voltage: a measurable but fleeting potential due to stray capacitance or EMI, not a functional fault. The difference between your initial high-impedance test (full voltage) and Low-Z (8V) is expected—high-impedance meters amplify these effects, while Low-Z mitigates them.
Likely Cause
The 8V is most likely caused by capacitive coupling between the BMS balance leads or battery wiring and the chassis, with the BMS’s activation reinforcing this effect. Here’s why:
The plastic sheets isolate the cells, but the wiring (especially the long balance leads for 16S) may run near the chassis, creating a capacitor-like effect.
The BMS (JK-PB2A16S20P) has active balancing and sensing circuits, which could induce small voltages or EMI when powered on, recharging the capacitance.
The consistent 8V (rather than random values) suggests a specific capacitive relationship tied to the system’s layout or a subset of cells (e.g., 8V could loosely correlate to a fraction of the pack, though not exactly).
Diagnostic Steps to Confirm
Measure Negative Terminal to Chassis:
Test voltage in Low-Z mode between the battery’s negative terminal and the chassis. If you see a similar 8V (or ~43V, i.e., 51.2V - 8V), it could indicate the chassis is floating at a potential relative to the pack, with capacitance splitting the voltage.
Disconnect Balance Leads:
With the BMS off, disconnect all balance leads from the cells (leaving only B- connected if needed for safety). Measure chassis to positive again in Low-Z mode. If the 8V disappears, the balance leads are the culprit due to their proximity to the chassis.
Inspect Wiring Layout:
Check if balance leads or main cables run parallel to the chassis walls. Capacitance increases with proximity and wire length. Rerouting them away or adding extra insulation (e.g., sleeving) could reduce the effect.
Ground the Chassis:
Temporarily connect a wire from the chassis to the battery negative terminal (ensure no risk to BMS operation—check JK manual). Retest in Low-Z mode. If the 8V vanishes, the floating chassis was amplifying the issue.
Is This a Problem?
Functionally: This phantom voltage isn’t likely to affect your battery or BMS operation. The rapid drop in Low-Z mode and lack of sustained voltage suggest no short or leakage current that would drain the battery or cause heat/damage.
Safety: If the chassis remains isolated (no conductive path to positive), it’s not a shock hazard. However, in a humid environment or with external connections (e.g., to a vehicle frame), stray voltages could become an issue.
Peace of Mind: If it bothers you, grounding the chassis to battery negative or rerouting wiring should eliminate it.
Next Steps
Start with the negative-to-chassis test and balance lead disconnection to confirm the source. If it’s the balance leads, adjust their routing. If it’s the chassis floating, grounding it is a simple fix (just verify with JK BMS docs that this won’t interfere with the BMS—most standalone BMS units tolerate this). Let me know the results, and I’ll tailor the solution further!
Similar threads
