Milwaukee updated their popular M18 Fuel 2767 heavy duty cordless impact wrench, and there seems to be a problem with the new design.
Update 10/7/22: Milwaukee Tool provided a new statement, skip to it below.
Torque Test Channel received numerous complaints from users about their recently purchased Milwaukee 2767 impact wrenches, and how the tools were breaking unexpectedly.
Examination of the broken tools point towards the new components as the common mode of failure.
I have spoken to Milwaukee Tool – they are taking this very seriously and are investigating the matter.
In the meantime, I would advise holding off on purchasing the 2767 unless you are able to handpick a copy of the previous revision. Or, buy the 2767 from a retailer with an easy return policy, and keep your receipt in a safe place.
User Failure Reports
Torque Test Channel (TTC for short) is a popular YouTube channel that specializes in independent testing of fastening tools and related products. I trust their opinions and findings.
TTC shared with me – and now in public video (embedded at the end of this post) – that they had received an unusual number of user complaints about the typically well-regarded Milwaukee M18 Fuel 2767 cordless impact wrench.
Users reported two types of issues.
Some tools appear to suffer from intermittent power delivery when the trigger switch is first pressed, only in certain modes, and temporarily. From the footage I examined, the tools seem to stutter for up to a few seconds at low speeds before operating at full speed.
More significantly, there have been reports of mechanical failure. I was shown evidence of partial failure, which had resulted in diminished performance, and also complete failure where tools were rendered completely non-functional.
How the Gear Box was Updated
TTC discovered that Milwaukee updated the 2767 impact around August 2022.
The previous 2767 design has serial numbers with the prefix H96A. We’ll call this the “A” model or revision.
The newer revision has serial numbers with the prefix H96B, or “B”.
Looking closely at the A model’s drivetrain assembly, the cam shaft (12) houses the planet gears (14) in a one-piece cage, with the planet pins (15) slip-fit and supported on both ends. The rear end of the cam shaft is centered in the rear gear case (20) via a large bearing (17). A washer (16) between the gear cage and bearing retains the planet pins.
Looking at the B model’s drivetrain, the cam shaft (12) has the planet pins pre-installed via press-fit. The planet gears (14) slip onto the pins and are held in place via a large washer (69). A ring (68), described as the rear washer in the parts diagram, serves as a spacer in the rear gear case (20) and also a bushing.
A Closer Look at the Powertrain
From what TTC has seen with high repeatability, the planet pins are bending and breaking after high torque applications.
Model A: Planet gears are arranged within a cage structure, with the planet pins slipped through and supported on both ends.
Model B: Planet gears are slipped onto planet pins, which are fixed to the cam shaft on one end.
For both versions of the tool, the rotor is held concentric at the front via a bearing (part 29 in both diagrams above) that is seated at the rear of the gear case (not shown here), and at the rear via the end cap. The hammer assemblies are held concentric at the front housing via a bushing.
Model A: The hammer assembly is radially supported at the rear with a large bearing. The planetary gears are positioned between front and rear radial supports.
Model B: The hammer assembly is radially supported at the rear with a bushing (rear washer). The planetary gears are positioned beyond the rear radial support.
The 2767’s “A” design’s powertrain is relatively standard for impact wrenches and in general. It’s not the only way a planetary gear box can be configured, but it’s tried-and-true. It is unclear what prompted Milwaukee to redesign it.
I have been looking at everything from planetary gear load distribution modeling to failure analysis, and there are so many things that can go wrong.
I would urge everyone to avoid jumping to any assumptions.
If I had to guess, I would suspect an unaccounted-for vibrational force (possibly from the bushing placement) that is increasing the load on the planet pins well beyond design specs.
Whatever the cause, this seems like the type of problem that no amount of modeling, simulation, or in-house testing could have predicted or revealed. But, it’s showing up as users put the revised 2767 impacts to work, and that’s a problem.
I would give Milwaukee engineers the benefit of the doubt; they could have taken every precaution to ensure the B model performed as well as the A model under any and all conditions. But, even with such efforts, things won’t always go according to plan.
It’s unfortunate, but these things happen.
Kudos to Torque Test Channel
TTC put a lot of work into investigating, replicating, and sharing about the Rev B models in a responsible and respectable manner. They went above-and-beyond to serve and protect the interests of their audience and end users.
I am sure that the issue would have been noticed and investigated as users and service centers reported tool failures to Milwaukee.
But from what I can tell, TTC’s quick, diligent, and thorough attention to the matter alerted Milwaukee Tool and end users about it earlier.
Milwaukee Tool is Investigating
Milwaukee is investigating the issue, and provided the following statement:
We take all feedback seriously, and our teams are actively investigating this claim. If anyone in your audience is experiencing dissatisfaction, please send them our way! They have a couple of options…
The first is to return the tool to us for repair utilizing our eService portal. This allows them to send it to us, free of charge, to be evaluated by Milwaukee technicians to determine if their repair is covered under warranty or to provide an estimated max repair price. All tools are repaired with genuine Milwaukee parts and the average turnaround time is about 7-10 business days.
They can also have their tool repaired through their nearest authorized service. They can find their nearest center here. Lastly, they can always give us a call at 1-800-SAWDUST from 8:00am to 6:00pm CT, Monday-Friday.
ToolGuyd Opinion and Recommendations
If you have a “B” model, what have your experiences been like?
In my opinion, given what I have learned:
If you’re buying the 2767 in person, check the serial number on the outside of the box – you want the version with serial number starting with H96A.
If you buy the 2767 with serial number starting with H96B, keep your receipt handy. The issue might only affect some but not all Rev B models – nobody knows just yet why these tools are breaking, only how.
If ordering online, smaller independent retailers might be able to check serial numbers in the warehouse.
If your tool fails after the return period, consult Milwaukee’s advice, above.
My local Home Depot had a promotional display with plenty of copies of the impact wrench. All but one were the previous A model. There’s a sticker on each box with the serial number.
If you want the A model, they’re still out there.
I’ll provide updates as I learn more.
Torque Test Channel’s Video
Milwaukee Tool Update
Milwaukee Tool provided a statement on 10/7/2022. I have added line breaks for easier readability. Here is what they said:
We have recently learned that a number of users are experiencing issues with their 2767-20 ½” High Torque Impact Wrench w/ Friction Ring, and quickly launched an in-depth investigation to understand the claims.
Upon investigation, 2 distinct issues were uncovered:
1) We recently implemented a new mechanical design in the model 2767-20. This new design is distinguished by “H96B” in the serial code.
This new design has not held up to the Milwaukee name and will no longer be sold.
Our developmental testing of this mechanism did not reveal the symptoms described by users, but the feedback is clear, and our decision to replace product is ultimately the right thing to do for them.
We will continue to produce the original design that has been on the market for years, that is distinguished by “H96A” in the serial code. We will immediately stop production of the new design distinguished by “H96B” in the serial code.
2) There are also variances in electrical tolerances that may cause some tools to “stutter”. This was exacerbated by the new mechanical design in 2767-20 inventory that is distinguished by “H96B” in the serial code.
If users are experiencing an issue, they are encouraged to send their tool in for repair or replacement by utilizing our eService portal. All products with the “H96B” distinction on the serial code will be replaced with “H96A”.
Frequently Asked Questions
If I wanted to purchase the 2767, but am hesitant to do so until inventory with the original design is available, what should I do?
There is currently a mix of “H96A” and “H96B” serial codes for the 2767-20 in inventory at authorized Milwaukee Distributors.
The ONLY product affected is the 2767-20 in bare tool packaging distinguished by “H96B” in the serial code. These products are being sent back to Milwaukee Tool and will be replaced with inventory of “H96A” by November 1st.
Why did you make the design change at all? Why mess with a good thing?
Products can change several times within their lifetime for a variety of reasons, but the goal is always to exceed the users’ expectations. This time we clearly did not hit the mark.
Will the new tool have the same level of performance as the original?
Yes, we are reverting back to the original “H96A” mechanical design that has been on the market for years.
Why didn’t you catch this before it went to market?
A thorough investigation is in progress. We applied the exact same test parameters for both the original and the new design.
[End of Statement]