As mentioned a few weeks ago, we are conducting comparative testing on oscillating multi-tool blades at Bosch’s request and funding. Here’s the project overview and a discussion of the tools and equipment being used for the study.
The objective of the project is to see compare 3 oscillating multi-tool metal-cutting blades in several different areas.
Well, I ran into a few snags.
Advertisement
The whole point of setting up a testing fixture is the ability to remove all sources of potential bias and error. It took a few tries, but we’re back on track.
Multi-Tool Performance Behavior
No-Load Power Draw
When a multi-tool is turned on, there will sometimes be a power spike, or inrush current. Even in the absence of a spike, the tool typically requires a short time until its full set speed is reached. Recovery is typically fairly quick and the power draw stabilizes and becomes consistent.
Without any load applied, meaning the tool is on but not actively making any cuts, power draw is about 1.21 A for the Bosch multi-tool at its lowest speed.
No-load power draw is higher at higher speeds, and lower at lower speeds.
The bad news is that different tools behave similarly but differently enough where comparison isn’t really fair. The good news is there was enough usable data, although not quite consistent enough for time-of-cut and ease-of-cut use, to show that multi-tools blades seem to work equally well regardless of the tools they’re used in.
Advertisement
A Note About these Plots
The current sensor I am using outputs 0.1 volt AC for every 1.0 amp AC that it measures. I am using a 10X multiplier loop between the power cord and power outlet, meaning the sensor will read 10 A for every 1.0 A of actual current. It will therefore output a voltage equal to the number of amps of power draw. In other words, the charts shown here are actually plots of AC voltage of the sensor output vs. time, but since 1.0 A of current is measured as 1.0 V, the units can be interchanged. I just wanted to point out that this is not actually a plot of measured current vs. time, but can be used as a plot of current vs. time.
Typical Power Draw During Cutting/Loading
There is usually a short delay after which a tool’s power draw adjusts to an applied load. My understanding is that the slightly delayed increase in power draw is due to the tool’s electronic regulation that maintains speed under load. This timing seems to be repeatable, and so time-of-cut measurements can be made from either onset of load or onset of stable power draw, as long as they’re consistently taken from the same place.
When the tool is under load, some power fluctuations are to be expected. Here, the power draw under load is about 1.735 A ± 30 mA, and it’s fairly consistent.
Abnormal or Inconsistent Power Draw
Here, you can see the power-on spike, a short time during which the motor was allowed to recover, a fluctuating power draw, and then a drop to lower power draw level.
This is an example of a really bad run. During the first half, the multi-tool was cutting a nail, but the blade also grabbed it a little, forcing the entire tool to bounce back a little bit a few times. Then, during the second half, the blade simply grabbed the nail and spun it back and forth.
Test Materials and Clamping
The first tests involved 16D nails embedded in 1x-thick wood. The nails were tightly driven into the wood, but vibrations led to some nails rotating in their holes, a very undesirable effect that means unusable data.
I tried using 2x square wood stock, but reaching embedded nails meant going through the deeper wood. Different blades with different geometries means the potential for inaccurate conclusions.
So then I tried drywall screws (3″ overall length). And then Spax self-drilling screws (#8 x 2-1/2″ long, partially threaded). And then copper tubing. And then 1/4″-20 stainless steel threaded rod.
Each test material provided different challenges. The screws led to greater vibrations, which means lower cutting performance. The copper tubing wasn’t really taxing enough on the blades and was difficult to clamp. The threaded rod was stainless steel (oops), leading to very slow cutting times, vibration, and quickly-dulled teeth.
Although not quite successful, cutting threaded rod required a different test jig setup than for the other materials. This allowed me to change things up a little bit, and I’m going to back to cutting nails.
Instead of driving nails into wood and clamping the wood to the testing jig, I will be clamping a 2″ precision screwless vise to the jig, and will use the V-grooved jaws to securely and rigidly position 16D nails for cutting.
I’m waiting for more blades and some 2-1/4″ socket head cap screws for proper securing of the vise, but dry runs have shown that the screwless vise is the way to go. (Hoorah!)
Simplification
For the retesting round, which should be the last now that the testing fixture is redesigned and producing more consistent results, I will be focusing on using just the Bosch MX30 for testing. Each test blade will cut (20) 16D bright common nails as originally planned.
In the planning stages of the project, I had thought that testing with different brands’ multi-tools might help eliminate the potential for bias. In theory, one brand’s blades might be designed to work best with that brand’s multi-tools. In reality, there doesn’t seem to be any tool-blade bias, but there are many new factors introduced when comparing different tools with the same blade. That ends up being a test to compare multi-tools, which doesn’t seem to be needed right now. Maybe that will come next, but I’ve removed it from the blade comparison study.
The final testing setup is a little more complex than it needs to be, but it’s about as consistent as it could be. Right now a weight system is used to ensure even cutting pressure (~8 lbs), and more weight can be added as needed. The only way to improve upon this is to use pneumatic cylinders, but that doesn’t seem to be necessary and could introduce other unforeseen issues.
Table of Contents
Part 1: Comparison Overview
Part 2: Testing & Measurement Tools
Part 3: Test Setup Optimization (Current Page)
Part 4: Final Test Setup
Part 5: Best Cutting Speed
Part 6: Cutting Performance
Part 7: Durability
Jerry
Wow, talk about scientific. I’d probably just hammer 30 nails into a 2X4 and measure how long it takes to cut them,and repeat the test until the blade was ‘dull’. I’d do this with 3-4 blades of every make tested. I’m sure your test will be far more scientific, which is probably why you are the one doing the test instead of me.
Tim
Maybe you covered this previously, but it strikes me as odd that Bosch would commission independent testing from a blogger (no offense intended) when they’ve likely already got the resources to do it in-house, and could simply bring in an independent testing lab to certify everything. A previous employer of mine was a major manufacturer of recip blades with a substantially smaller R&D budget than Bosch, and we had test rigs with full PLC control, datalogging of every parameter, etc. We ran cut tests pretty frequently, so we were not “figuring it out as we went”. We’d routinely bring in a 3rd party testing lab to conduct trials with competitors products and have them certify that it was done uniformly. I’m certainly not knocking your ability to get valid results, and I’m sure it’s a cool project for you . . . it’s just strange to me. I’d have to imagine that Bosch has the equipment and personnel to do these trials themselves and independently certify them, so why farm it out unless you are interested primarily in the page views it’ll generate?
Stuart
The arrangement includes full disclosure of results and conclusions with readers, no matter what the findings are. I approached this with a blank unassuming slate, but it would be foolish not to guess that Bosch hasn’t already seen that their blades best competitors’.
I’m sure they have done full testing themselves and might even have one or more engineers or techs running continued testing.
Bosch was fine with me conducting simple handheld tests. But I don’t like handheld tests as they can be easily skewed.
I don’t often conduct comparative tests because they must be incredibly involved for me to be 100% confident in the findings. The complexity is my own choice. If I’m going to compare competing brands’ blades, I want to be absolutely sure of any conclusions. There must be no doubt, and I want to be able to show exactly what I base my conclusions on.
As you guessed, this is a really cool project for me. Comparing the blades’ performance in terms of the tool’s power draw is not something I would have been able to budget time or resources for as part of regular ToolGuyd content.
Tim
I gotcha. So this is more of a science project for you, one which likely Bosch feels confidant enough in their offering to bankroll, and not an effort by Bosch to get quantitative numbers in return? Again, not knocking the efforts or methodologies, but if they came out as top dog by your metrics . . .legally it’d be a reach for them to use those findings to make claims of superiority in any marketing efforts. It’s not really like they are “sourcing” the testing to you, just allowing you to indulge your own curiosities for the benefit of readers (and some product exposure)?
Jeremy
Tim: Are there any manufacturers who don’t claim they have made the best blades in some category or other? I can’t think of any offhand who won’t claim that they make blades that last 2x longer than the competition, cut more, cut faster, smoother etc. None of them seem to publish the actual data(or comparisons) in an easy to read document anywhere that I know of.
While I’m sure that legally they are right for marketing purposes, and have the independent lab tests to prove just how right they are, it doesn’t mean much to me as an end user who just wants to know in a general way which blade will cut more quickly in my type of use. I have no faith that the independent lab tests will actually relate to my usage in any meaningful way.
Tim
Understood and largely agree with all that. Certainly not trying to knock or discredit the efforts or motives on your end. I was just surprised to see a manufacturer financially backing a privateer effort to do competitive testing with it’s products.