A parasitic draw is current the car keeps pulling after you lock it and walk away. Finding one is a measurement problem before it is a repair problem, and most people measure it wrong.
ModerateSafe to drive$85–$1,100 diagnosis and repair typical repair
Quick answer
Every modern car draws a small amount of current with the key off to keep clocks, alarms, keyless entry and module memory alive. That is normal and it is called parasitic draw, dark current or key-off current. It becomes a fault when it is large enough to flatten the battery while the car sits. There is no single acceptable milliamp figure — published limits run from 7 mA to 100 mA depending on who you ask, because each manufacturer sets its own limit against its own battery's capacity. Subaru's 2025 service bulletin says under 70 mA; Fluke and Snap-on say under 50 mA; OPTIMA says 25 mA. The more useful rule is capacity-relative: divide the battery's reserve capacity in minutes by 4 to get your own maximum in milliamps. And before you measure anything, the modules have to go to sleep — that takes 20 minutes on most cars and up to several hours on some, so a reading taken immediately after locking the doors will look like a catastrophic fault when the car is perfectly healthy.
Diagnostic summary
Problem
Parasitic Draw
Severity
ModerateGet it looked at soon
Can you drive?
Safe to drive
A parasitic draw does not affect how the car drives. Nothing about the steering, braking or engine changes. What it takes away is the ability to leave the car parked — the failure always happens at the start of a trip, never during one. Plan around being stranded rather than around a breakdown, and carry a jump pack until it is fixed.
Typical repair
$85–$1,100 diagnosis and repair
Common causes
Aftermarket electronics wired to constant power, A control module that never goes to sleep, A door, hood, trunk or glovebox switch that never opens
Symptoms you might notice
Battery is flat after the car sits, but fine with daily use
This is the defining pattern. Subaru's service bulletin requires a verifiable history of repeat dead batteries after the vehicle has been sitting — overnight, over a weekend, or longer — before a technician is authorised to run the draw test at all. A battery that dies while you are driving is a charging-system problem instead.
A consistent number of days before it goes flat
Owners usually describe either 'dead every single morning' or 'dead if I leave it more than a week'. That distinction is diagnostic, not incidental: a draw over roughly 500 mA flattens a battery overnight, while a draw under 100 mA takes one to three weeks. The complaint tells you roughly what magnitude to expect on the meter.
Repeated jump starts, each one working normally
The car starts and runs perfectly once jumped, which rules out a starter or ignition fault. The battery is being emptied, not damaged in a way that stops it working.
A replacement battery died the same way
The single strongest indicator. If a new battery repeated the pattern within weeks or months, nothing was ever actually fixed. Repeated deep discharge also sulfates the new battery, so by the time the draw is found you may need a third one.
No warning light while driving
The dashboard tells you nothing, because there is no lamp that monitors key-off current. The battery light reports charging-system voltage while the engine is running, and it stays off during a parasitic draw. Its absence is a clue, not a reassurance.
Symptoms started after something was installed
A dash cam, GPS tracker, alarm, remote start, aftermarket stereo or a permanently plugged-in OBD-II dongle. If the timeline lines up with an installation, the timeline is usually right.
What causes it
1
Aftermarket electronics wired to constant power
Common
Dash cameras, GPS trackers, alarms, remote-start systems and plug-in OBD-II dongles are frequently wired to a permanently live circuit so they keep working while the car is parked — which is exactly the point of them, and exactly the problem. Subaru's 2025 bulletin instructs technicians to disconnect these before testing at all, noting that many draw power continuously. The trap is that each device may draw an amount that is individually acceptable and still push the total over the limit; Subaru specifically warns that draw is cumulative.
How to spot it
The problem began after an installation, or the draw drops when the accessory's fuse is pulled or its plug is unseated.
The fix
Move the device to a switched (ignition-live) circuit, add a hardwire kit with a low-voltage cutoff, or unplug it when parking for more than a day or two.
$0–$250•Moderate DIY
2
A control module that never goes to sleep
Common
Body control modules, infotainment head units and telematics modules are supposed to power down in stages after you lock the car. One that stays awake keeps drawing the current it uses while running — Snap-on puts a stuck-awake module as high as 150 mA. Subaru maintains a dedicated escalation procedure specifically for its Data Communications Module, the telematics unit, which is a strong signal that connected-car modules are a real-world source on that platform, though no manufacturer publishes frequency data.
How to spot it
The draw is steady and stays high well past the normal sleep window, and it tracks to one module's fuse rather than a lighting or accessory circuit.
The fix
Confirm the circuit with a fuse pull, then check for a software update or service bulletin covering that module before condemning it. Replacement usually requires programming.
No reliable published range•Shop recommended
3
A door, hood, trunk or glovebox switch that never opens
Common
Latch and pin switches tell the body module whether the car is closed up. A misadjusted or corroded one reports a door permanently ajar, so the module never sleeps and an interior lamp may also stay lit. Subaru names faulty or misadjusted door-pin, hood-latch and trunk-latch switches as a cause of a permanently awake Body Integrated Unit; a glovebox or trunk bulb that never goes out is the older, simpler version of the same fault.
How to spot it
An interior light stays on, a door-ajar indicator behaves oddly, or you can feel warmth from a trunk or glovebox bulb after the car has been parked.
The fix
Adjust or replace the offending switch. Confirm by opening the boot or glovebox and checking the lamp actually extinguishes when the lid closes.
$15–$250•Moderate DIY
4
Damaged wiring, corroded connectors or water intrusion
Common
Chafed insulation, a broken conductor and moisture in a connector all create current paths that should not exist, and they are the hardest kind to find because they are often intermittent. MOTOR documented a case where a power-supply wire had broken inside the rubber boot between a door and the body: a slight movement of the door woke the body control module. The measured draw when it did occur was 8.79 A — enough to flatten a battery in a single night — but it only appeared some of the time.
How to spot it
The draw comes and goes, changes when a door or harness is moved, or appears only after rain or a car wash.
The fix
Trace the circuit with the wiring diagram, flexing harnesses and boots while watching the meter. Repair the conductor and seal the entry point rather than just taping it.
$100–$600•Advanced DIY
5
A relay stuck closed
Occasional
A relay whose contacts weld or whose coil circuit is shorted keeps a high-current load energised with the key off. Because relays feed things like fuel pumps, blowers and cooling fans, this failure tends to produce a large draw rather than a marginal one, and the relay itself is often warm to the touch after the car has sat.
How to spot it
A draw in the amps rather than milliamps, and a relay that is noticeably warm or that clicks when tapped.
The fix
Swap the suspect relay with an identical one from a non-critical circuit and re-measure. Replacement relays are inexpensive.
$20–$200•Moderate DIY
6
A worn alternator diode
Occasional
The alternator's rectifier diodes are supposed to pass current one way only. A failed diode lets the battery discharge backwards through the alternator when the engine is off. This one matters out of proportion to how often it happens, because it is the single failure that produces a genuine parasitic draw while also being a charging-system fault — so it can look like either problem depending on which test you run first.
How to spot it
The draw disappears when the alternator's main output lead or its fuse is disconnected, and charging output may also be marginal or produce AC ripple.
The fix
Confirm with an AC-ripple measurement at the battery while running, then replace the alternator. See alternator replacement costs.
$500–$1,100 installed•Advanced DIY
7
A key fob left inside the vehicle
Occasional
Keyless-entry systems poll for a fob, and a fob left in the cabin or in a bag inside the car can keep the receiving module awake indefinitely. Subaru names this explicitly, both as a cause of repeat dead batteries and as a test error, and requires fobs to be kept at least 4.92 feet (1.5 metres) away during measurement. It is worth checking before anything else because the fix is free.
How to spot it
A spare fob lives in the glovebox or a bag that stays in the car, and the draw disappears when it is removed from the vehicle.
The fix
Keep spare fobs out of the vehicle. If the car has a valet or transport mode that disables polling, use it for long-term parking.
$0•DIY friendly
How to diagnose it
1
Rule out everything else first
Subaru's bulletin is unusually explicit that a draw test should only be performed after other causes of a dead battery have been eliminated — a battery confirmed good by a load test, a charging system confirmed good, obvious errors like a dome light or accessory charger ruled out, and a short-trip driving pattern ruled out. This is not procedural politeness. A sulfated battery and a 300 mA draw produce an identical complaint, and hunting a draw on a car with a dying battery wastes hours and finds nothing.
Result: Prevents the most expensive mistake on this topic: diagnosing a draw that does not exist.
2
Ask whether it dies from sitting or from use
This single question splits the field. Dies after sitting means a draw or a battery losing capacity. Dies while driving or immediately after a long drive means the charging system. Dies only in cold weather but recovers means a weak battery. Dies only after a run of short journeys means the alternator never had time to replace what starting consumed. Only the first pattern justifies going further.
Result: Sorts a parasitic draw from the three faults that imitate it.
3
Set the car up so it does not wake back up
Close and latch every door, the hood and the trunk — latch them, do not just close them, or the latch switches keep the body module awake. Subaru requires key fobs to be at least 4.92 feet (1.5 metres) from the vehicle to prevent polling. ALLDATA suggests taping down the dome, glovebox, under-hood and trunk light switches. Disconnect any aftermarket electronics, and unplug any scan tool from the diagnostic connector — Snap-on notes a tool left on Pin 16 is itself a load.
Tools: Tape for latch switches, Somewhere to put the key fobs
Result: Ensures the reading you eventually take describes the car, not your presence next to it.
4
Connect the meter without blowing its fuse
Set the multimeter to DC amps with the red lead in the 10 A port, disconnect the negative battery cable, and bridge the gap with the meter so all current flows through it. The classic mistake is forgetting and cranking the engine: AAMCO's training manual describes the starter trying to source 200 amps through the meter's half-amp fuse. Protect against it with a fused test lead — AutoLab suggests 5 to 8 A — and never substitute a larger fuse than specified. The safest version of this setup is Clore's: fit a high-current switch into the negative cable and read across the switch, so the circuit is never actually broken.
Tools: Digital multimeter with a 10 A range, Fused test lead, 5–8 A, Optional: switch-in-cable adapter
Result: Gets a total key-off current reading without damaging the meter or resetting the car.
5
Wait for the modules to sleep — this is the whole test
Snap-on reports that a normal vehicle is still drawing roughly 500 milliamps ninety seconds after shutdown. Read at that moment and you will conclude the car has a severe fault. Subaru states the power-down process takes a minimum of 20 minutes on most models and warns in as many words that a premature reading gives a false indication of excessive draw, leading to incorrect and time-consuming repairs. Published full-sleep times run from Snap-on's 10 minutes to Clore's observation that an air-conditioning control module can stay awake up to six and a half hours. Watch the reading decay and stop when it goes flat rather than working to a fixed clock.
Tools: A meter you can leave connected, Time — 20 minutes minimum, potentially hours
Result: Separates the normal shutdown sequence from an actual fault. Skipping it invalidates everything after it.
6
Compare the reading against your own battery, not a generic number
Rather than arguing about whether 50 or 70 milliamps is the limit, use the capacity-relative rule that Clore and Vehicle Service Pros both describe: divide the reserve capacity in minutes by 4, or the amp-hour rating by 2.4, to get the maximum acceptable draw in milliamps. A 100-minute reserve capacity gives 25 mA; a 120-minute battery gives 30 mA. The number is printed on the battery label. This is the only published rule with engineering logic rather than habit behind it.
Tools: The specification label on your battery
Result: Produces a defensible threshold for your specific vehicle instead of a borrowed rule of thumb.
7
Pull fuses one at a time — and do not put them back
With the meter still reading, remove a single fuse and watch for the current to fall. Subaru's procedure is strict about three things, and each one exists because technicians get it wrong: always remove one fuse at a time; wait five minutes or more after each pull for the value to stabilise; and do not re-insert the fuses, because replacing one can wake a module, re-establish the draw and force the whole waiting period to start again. Record the value at every pull rather than only the one that drops it, because draw may be cumulative across several circuits.
Tools: Fuse puller, Wiring diagram, Something to write the readings on
Result: Narrows a whole-vehicle draw to a specific circuit, or reveals that it is spread across several.
8
If no fuse drops it, suspect a network that never slept
Subaru describes a failure mode almost nobody else publishes: if the draw stays high after the wait period and no single fuse pull reduces it, the communication network itself may never have gone to sleep — in which case fuse-by-fuse diagnosis is invalid until that is resolved. The named causes are a key fob left inside, a faulty or misadjusted door-pin, hood-latch or trunk-latch switch keeping the body module awake, or a module continuously transmitting on the network. Chasing individual circuits at this point finds nothing, which is exactly why this diagnosis runs up hours.
Result: Explains the frustrating case where the draw is real and no circuit accounts for it.
9
Use voltage drop or an amp clamp when disconnecting is the problem
Both alternatives exist because breaking the circuit can hide the fault — Subaru notes that disconnecting the battery may itself temporarily fix the issue by sending the system to sleep along with the error. Measuring millivolt drop across each fuse in place disturbs nothing: Subaru gives the calibration example that a 10 A fuse passing 175 mA shows about 1.5 mV, against 0.1 to 0.2 mV with no flow. An inductive amp clamp is the other non-intrusive option, but resolution decides whether it works — AAMCO says you need one that reads down to 2 to 3 milliamps, and warns that large-jaw clamps sized for battery cables are not suitable for this job.
Tools: Multimeter with millivolt resolution, Low-current inductive amp clamp (1–3 mA resolution)
Result: Finds intermittent draws that vanish the moment a conventional test setup is connected.
Repair options & cost
Shop diagnosis (billed as diagnostic time)
There is no flat-rate figure for this job because it is a hunt, not a defined repair. It is billed at the shop's hourly rate — a 2026 survey of over 10,000 North American shops puts the national average at $132 per hour, ranging from about $85 in the cheapest states to $197 in the most expensive, and AAA reports dealers typically charge 25 percent or more above independents. One hour is a common minimum. Ask for a written diagnostic cap and a stop-and-call point before authorising open-ended time.
$85–$600Shop recommended2 hrs labor
DIY measurement with a multimeter
The equipment cost is modest and the test is within reach of a careful DIYer. What it actually costs is patience: the waiting period before a valid reading is 20 minutes at best and potentially hours, and Subaru's procedure requires another five-minute settle after every fuse pull. Budget an afternoon rather than an hour.
$25–$120 tools onlyModerate DIY3 hrs labor
Repairing the circuit once it is found
Deliberately no range here, because the honest answer is that it depends entirely on the culprit and the spread is enormous. A misadjusted door-pin switch and a body control module are three orders of magnitude apart in price. Clore's published case study on a 2002 GMC Envoy is representative of how the repair itself is often the small part: an initial 0.24 A fell to 0.03 A once the offending fuse was identified, and settled at 20 mA after the module was replaced.
No reliable published rangeShop recommended
Battery replacement after repeated deep discharge
Frequently necessary alongside the real repair, not instead of it. Lead-acid batteries sulfate when left discharged, and a battery that has been flattened repeatedly by a draw has usually lost capacity permanently. Load test it after the draw is fixed rather than assuming it recovered. See battery replacement costs.
$120–$320 installedDIY friendly0.5 hrs labor
A smart maintainer as a stopgap
Not a fix, and worth saying so plainly — it masks the symptom while the draw continues. But it keeps a car usable while you wait for a diagnostic appointment, and it protects a healthy battery from the sulfation damage that repeated deep discharge causes.
$30–$90 toolDIY friendly
No source we could find publishes a price for parasitic draw diagnosis specifically, and we are not going to invent one. RepairPal lists $122 to $179 for generic electrical system diagnosis, labor only, with no stated labor hours and no publication date. What can be stated honestly is the structure: parts cost nothing for the diagnosis itself, labor is billed as diagnostic time at $85 to $197 per hour depending on state, and the hours vary more than for almost any other job because the waiting periods are long and an intermittent draw may not present during the visit at all. In MOTOR's documented case, the customer's battery only died if the car sat three days or more, and even then only about half the time.
When to see a mechanic
▸You measure a draw over 1 amp. That is not a memory circuit, it is something substantial staying energised, and it can flatten a battery overnight.
▸You smell burning plastic, find melted insulation, or a fuse box that is warm to the touch. Stop testing and have it inspected — this is a fire risk, not a battery inconvenience.
▸The draw is real but no fuse pull reduces it. The network may never be sleeping, and fuse-by-fuse diagnosis cannot resolve that.
▸The circuit involves airbags or restraints. SRS wiring should not be probed or disconnected without the correct depowering procedure.
▸The vehicle is a hybrid or EV. High-voltage systems have their own key-off behaviour and their own hazards, and the twelve-volt battery is charged by a DC-DC converter rather than an alternator.
▸The battery case is swollen, cracked, leaking or hot. Do not charge it and do not jump it.
▸You have replaced the battery twice and it keeps happening. Two batteries is the point at which further guessing costs more than a diagnosis.
What counts as a normal parasitic draw
This is the question everyone arrives with, and it does not have the answer they want. Below is what published sources actually say, unaveraged and unreconciled, because averaging them would manufacture a consensus that does not exist.
Source
Acceptable
Definite fault
Notes
Subaru of America, SIB 07-236-25 (2025)
Under 70 mA
—
Primary manufacturer document; allows slightly higher on models with more electronics
Snap-on
Under 50 mA
150 mA cited as a stuck-awake module
Undated
Fluke (2025)
Under 50 mA
Over 100 mA
Test-equipment maker
OPTIMA Batteries (2022)
25 mA
Over 100 mA
Vehicle era unspecified
Universal Technical Institute (2021)
50–85 mA newer, under 50 mA older
—
The origin of a much-copied line
Interstate Batteries (2022)
20–50 mA, more if newer than 2009
—
Battery manufacturer
ALLDATA (2023)
Under 50 mA per circuit
—
Per-circuit, not total — a different measurement
Clore Automotive
Reserve capacity ÷ 4
—
Capacity-relative rather than fixed
Vehicle Service Pros (2022)
Under 75 mA during shutdown
—
Trade article; also lists per-brand figures we could not verify
AAMCO training manual LBT-270
7–12 mA typical, 30 mA rule of thumb
About 150 mA
Also quotes 15–40 mA elsewhere in the same document
Identifix
50–100 mA described as typical
Over 1 A
Undated; see the caution below
Published limits for acceptable key-off current
The spread for acceptable runs from 7 mA to 100 mA — a fourteen-fold range — and the threshold for a definite fault runs from 100 mA to a full amp. This is not sloppiness on anyone's part. Each manufacturer sets its own limit against its own battery's reserve capacity and its own module architecture, so there is no cross-brand number to give.
Figures circulate widely for specific brands — Toyota 50 mA, Ford 50 or 80 mA, GM 40 mA, Nissan 25 mA, Honda 27 to 42 mA, Dodge and Ram 5 to 35 mA. We could not retrieve a primary manufacturer document for any of them; they trace back to a single 2022 trade article and to forum posts. The only manufacturer figure verified here against the manufacturer's own published document is Subaru's 70 mA. Treat the rest as hearsay until you see them in your vehicle's own service information.
Why measuring too early invents a fault that is not there
If there is one thing on this page that saves someone a wasted weekend, it is this. Locking a modern car does not switch it off. Modules run self-tests, write data and power down in stages, and until that finishes the car legitimately draws hundreds of milliamps. Snap-on gives the number that makes it concrete: ninety seconds after shutdown, a normal vehicle is still pulling roughly 500 milliamps. Measure then and you will diagnose a catastrophic drain on a perfectly healthy car.
Source
Stated time
Snap-on
Normally fully shut down within 10 minutes; BMW forced sleep mode completes in 3 minutes 45 seconds
Subaru, SIB 07-236-25
Minimum of 20 minutes on most models; appendix cites a 25-minute minimum wait
Fluke
10 to 45 minutes
ALLDATA
30 to 120 minutes
Vehicle Service Pros
About 10 minutes for the initial phase, up to 2 hours for full sleep
AutoLab Scope Diagnostics
20 minutes on older model years to 2 hours on newer; author allows a minimum of 2 hours on late models
AAMCO LBT-270
As little as 30 minutes, up to 4 hours
Clore Automotive
Several hours for some modules; an air-conditioning control module can stay awake up to 6.5 hours
How long modules take to sleep, by source
Three minutes forty-five seconds to six and a half hours. The variance is the story, and it is why this diagnosis costs what it costs — most of the billed time is waiting. The practical approach is not to work to a fixed clock but to watch the reading decay and stop when it stops falling.
How long a given draw takes to strand you
The arithmetic is simple, but one detail changes the answer by roughly half and most published versions leave it out. A lead-acid starter battery will generally not crank a cold engine much below about 50 percent state of charge. So the number you want is not time to fully flat, it is time to no-start — and that is roughly half of what a naive calculation gives.
Formula
hours to no-start = usable capacity in mAh ÷ draw in mA
Usable capacity
nominal amp-hours × 1000 × 0.5
Worked example
A 50 Ah battery has roughly 25,000 mAh usable before it stops cranking
Draw
Hours
In practice
30 mA
833
About 35 days
50 mA
500
About 21 days
70 mA
357
About 15 days
100 mA
250
About 10 days
300 mA
83
About 3 and a half days
500 mA
50
About 2 days
1 A
25
About a day
2 A
12.5
Overnight
Time to no-start on a 50 Ah battery (25,000 mAh usable)
The table also reconciles the two complaints owners bring in. A draw under 100 mA produces 'it dies if I leave it a week or two'. A draw over 500 mA produces 'it is dead every single morning'. Before you measure anything, your own description of the symptom already tells you roughly which order of magnitude to expect.
Battery capacity varies enough that the table is a shape rather than a prediction: group sizes in common use run roughly 40 to 70 amp-hours, and reserve capacity is typically 120 to 130 minutes on a group 48. Read your own battery's label and rerun the division. Nothing here beats the number printed on the part in your own engine bay.
When it is not a parasitic draw at all
Subaru's bulletin makes the draw test conditional, and the conditions are worth reproducing because they eliminate most cases before any equipment comes out. The test should only be performed once the following have been ruled out.
1
The battery has been tested and confirmed good
A sulfated or internally failing battery presents identically — fine when driven daily, dead after a weekend. Most US chain parts stores will load test it free in the parking lot. Note also that leaving a battery discharged for even a few hours causes permanent damage, so a battery that has been flattened repeatedly may need replacing regardless of what caused it.
2
The charging system has been tested and confirmed good
This is where the battery light is genuinely useful, and it is worth understanding why: that lamp reports charging-system output while the engine runs. It has nothing to do with key-off current. A car with a real parasitic draw typically shows no battery light at all.
3
The obvious has been ruled out
A dome light left on, a door not fully closed, a key fob left in the cabin, an accessory charger still plugged in. Subaru lists these by name because technicians find them regularly, and none of them require a diagnostic charge.
4
The driving pattern has been considered
Subaru's mandated customer interview asks how often the vehicle is driven and whether it is used only for short trips, because a car driven in short bursts may never run long enough to replace what starting consumed. Over months that becomes a genuinely weak battery. The cause is the schedule, not the wiring.
5
The failure pattern actually fits
Subaru requires a verifiable history of repeat dead batteries after the vehicle has been sitting. A battery that dies while you are driving, or right after a long drive, is not a parasitic draw and no amount of fuse pulling will find one.
What we deliberately did not tell you
Three things about this topic are worth stating openly, because the gap between what gets published and what is actually established is unusually wide here.
No source publishes a ranked list of causes. The likelihood labels on the causes above reflect how consistently technician-facing sources name each one, not measured frequency data, because no such data appears to be published. Anyone telling you the most common cause of a parasitic draw is a specific component is guessing with more confidence than the evidence supports.
No source publishes a price for this specific job. Every figure you will see quoted is either a generic electrical diagnosis charge or an hourly labor rate. We have given you the rate and the reason the hours are unpredictable instead of a fake-precise total.
The advice to avoid inrush current when connecting the meter is universal and unquantified. Every source repeats it; none we found gives a magnitude or a duration. The risk is real enough to design around — hence the switch-in-cable method — but we cannot tell you how large it is.
One more thing worth knowing before you visit a dealer: the OEM term for this is often dark current, not parasitic draw. Subaru uses it throughout its service documentation. They mean the same thing, and recognising it will save you a confusing conversation at the service desk.
It is the current a vehicle continues to consume after it is switched off and locked. A small amount is normal and necessary — it keeps the clock, the alarm, the keyless-entry receiver and the memory in various control modules alive. It becomes a fault only when it is large enough to flatten the battery while the car sits. Manufacturers usually call it dark current or key-off current in their own service documentation.
▸What is an acceptable parasitic draw in milliamps?
There is no single number, and anyone who gives you one is quoting one source out of many. Published limits run from 7 mA to 100 mA. Subaru's 2025 service bulletin says under 70 mA; Fluke and Snap-on say under 50 mA; OPTIMA says 25 mA. The rule that actually generalises is capacity-relative: divide your battery's reserve capacity in minutes by 4, or its amp-hours by 2.4, to get your own maximum in milliamps. A 120-minute battery gives you 30 mA.
▸How long do I have to wait before measuring?
At least 20 minutes, and potentially several hours. Subaru specifies a minimum of 20 minutes on most models; Fluke says 10 to 45; ALLDATA says 30 to 120; Clore notes an air-conditioning module can stay awake up to six and a half hours. This is the step people skip and it is the step that matters — Snap-on reports a normal car is still drawing about 500 milliamps ninety seconds after shutdown. Watch the reading decay and take your measurement when it stops falling.
▸Does the battery warning light come on with a parasitic draw?
No, and that is diagnostically useful rather than disappointing. The battery lamp monitors charging-system voltage while the engine is running. Nothing on the dashboard monitors key-off current, so a parasitic draw is silent. If the light is on while you drive, you are looking at an alternator or belt problem instead.
▸Can a parasitic draw ruin a new battery?
Yes, and this is why replacing the battery without finding the draw so often fails twice. Lead-acid batteries sulfate when they sit discharged, which permanently reduces capacity — damage begins within hours, not weeks. A new battery repeatedly flattened by an undiagnosed draw can lose meaningful capacity in months. Fix the draw first, then load test the battery before deciding whether it survived.
▸How much does it cost to find a parasitic draw?
It is billed as diagnostic time rather than as a fixed job, because it is a hunt. Shop labor rates averaged about $132 per hour nationally in 2026, ranging from roughly $85 to $197 depending on state, with dealers typically 25 percent or more above independents. One hour is a common minimum and two to three is not unusual, largely because the waiting periods are long. Ask for a written cap before authorising open-ended diagnostic time.
▸Can I test for a draw without disconnecting the battery?
Yes, and sometimes you have to. Disconnecting the battery can temporarily fix the symptom by sending the whole system to sleep along with whatever fault was keeping it awake. The two non-intrusive methods are measuring millivolt drop across each fuse in place — Subaru gives the example of a 10 A fuse passing 175 mA showing about 1.5 mV against 0.1 to 0.2 mV with no flow — and using an inductive amp clamp with enough resolution, which means one that reads down to 1 to 3 milliamps rather than a large-jaw clamp meant for battery cables.
▸Why did my reading not drop when I pulled any fuse?
Most likely the communication network never went to sleep, in which case fuse-by-fuse diagnosis cannot work until that is resolved. Subaru names three causes: a key fob left inside the vehicle, a faulty or misadjusted door-pin, hood-latch or trunk-latch switch keeping the body module awake, or a module continuously transmitting on the network. The other possibility is that the draw is spread across several circuits rather than concentrated in one, which is why you should record the reading at every fuse pull rather than only the one you expect to matter.