Developing Standards to Properly Rate Esk8 Components

#1

Lately a lot of manufacturers have been making power/amperage claims that have caused discussion in the community.

One of the problems is that there is no standardized procedure for measuring the capabilities of the equipment in esk8. This can cause a lot of confusion since there is no way of making a fair comparison between components.

Example:

  • Manufacturer A claims that his ESC can output 60A continuous @ 50V.

  • Manufacturer B claims that his ESC can output 80A continuous @50V.

So ESC from manufacturer B is more powerful right?

Unfortunately unless we know all the testing parameters there is no way of telling.

What is not apparent in the previous example is that A classifies as continuous operation 5min of constant output with the ESC screwed on a huge aluminium block with an ambient temperature of 20C while manufacturer B 3min of constant output with a fan blowing on the case with an ambient temperature of 15C.

There is literally no way of comparing these two speed controllers using the manufacturer provided specifications.

So why not establish a standard of measurement?

For example welding machines have their current ratings measured in combination with the duty cycle.

So you may have the following different 230A welding machines.

  1. 230A @35% , 130A @%100

  2. 230A @60%, 180A @100%

Both machines are sold as 230A welders however machine nr.2 is capable of sustaining its output for longer.

Duty cycle is measured in a 10min duration at 40C ambient temperature.

So 180A @ 60% duty cycle means the machine can operate at 180A for 6min, cool down for 4min and so on.

Why not agree on a standard procedure for rating components?

Obviously there will be manufacturers that will not want to do so, since this would mean they would lose the ability to use bullshit ratings as a marketing tool but in the end these manufacturers should not be trusted.

Nobody buys a car or a motorcycle on unbased manufacturer claims, why do accept this in esk8?

26 Likes
*SCAM ALERT* neoBOX open source Focbox variant (HW 4 and HW6 Design)
#8
#9

This is very much needed. When first investigating escs and vescs, continuous vs peak amps was very floaty and not scientifically approached as a standard.

It seems not much has changed.

8 Likes
#10

This is great, because i think we’re confusing a lot of people into thinking they need to shoot for these inflated values in order to have their builds be impressive or get noticed.

3 Likes
#11

To achieve these ratings, we could set up an independent testing facility (someone’s garage) and have each vendor who wants their product be “certified” send a sample there.

If a vendor refuses, we could still crowd fund the purchase of a unit.
This should save us more money than cost us in the long run.

8 Likes
#12

At the beginning this is not necessary.
As long as the testing procedure is standardized is would be easy to validate any vendor so there will be fewer false specs.
Manufacturer XYZ will think twice before marketing his motor as 100A capable when he knows it will be easily debunked.
If a vendor does not want to submit his product to the standard tests, it should not be the buyers that bear the cost of testing.

3 Likes
#13

I fear nothing will change as long as there is no independent authority actually validating claims.

5 Likes
#14

A bunch of DIY guys testing parts in the garage aren’t exactly an authority.

8 Likes
#15

I’d love to see an ESK8 dyno of sorts that puts an actual physical load on motors. It is completely doable and would be easy to conduct tests on a bunch of ESCs.

Plus the videos of dying MOSFETs will be cool. :joy:

4 Likes
#16

In my opinion establishing continuous current ratings should not require a complex workbench/setup and would essentially just need a variable current source at fixed voltage and a way to systematically increase the load on the ESC (which may be easier said than done).

A simple test could be increasing the load in steps at 100% duty cycle so the current ups 5 amps every 15 minutes on an air-suspended ESC until failure occurs. You want to give the ESC enough time to reach thermal equilibrium, 15 min blocks should be enough time. Of course failure may occur for these semiconductor parts along with an absurdly hot temperature, but this is still a great indicator of the actual limits of just the ESC itself and would allow for pretty much all ESCs to be tested and ranked the same way. Unfortunately this type of test would destroy the ESC so it is unlikely that hobbyists themselves would want to attempt it unless they are not worried about breaking these.

ESCs should be tested suspended in open air at at ambient temp (23 degree celcius) with no forced air cooling. Using any type of heatsink apart from what is attached to the ESC itself could mean unfair advantages/disadvantages for ESCs that are able to mate with the heatsink more or less efficiently, which could occur just from the surface structure or shape of the ESC heatsink (or lack thereof). Even the factory plastic heatshrink could make a dramatic difference in heat dissipation in air, so I would recommend that every test completely remove the heatshrink wrap from the ESC. While a heatsink/increasing cooling would certainly increase the current rating, it just confounds the current comparison to other ESCs.

It would be best to have a 3rd party test this, as all manufacturers have a conflict of interest in testing their own equipment. While a DIY enthusiast in their garage may not be an authority, a DIY enthusiast with the proper testing setup and properly documenting each test should have no problem creating a test that accurately compares ESC ratings and may be a very credible reference.

Maybe the most important point is that whatever the test is, it needs to make a fair comparison between each ESC eliminating advantages/disadvantages from geometry, heatsinks, or other discrepancies between ESCs. I think this is actually quite hard to do, as some ESCs come built with geometry or heatsinks alone that will certainly increase their ratings aside from the actual electronic hardware itself.

5 Likes
#17

I wasn’t referring to a bunch. I was referring to a single person with a good setup.

#18

Would you like to be that third party? ^^

#19

My comment still stands. Regardless if it’s one person or two or three or 10

Trying to create an authority on ESC testing just sounds silly to me.

What doesn’t sound silly is having a discussion to try and convince manufacturers to rate their things based on the same standards

1 Like
#20

I think that if the components come with said heatsinks/geometry as standard, why should that not be incorporated in the test?

2 Likes
#21

What do you think how certifications/testing labs started? All that is needed is a man on a mission to get it going.

Authority was the wrong word. I mean a trusted source, that influences buyers decisions.

1 Like
#22

I’m all for having someone do unbiased reviews and testing with real world numbers.
The word authority is what lost it for me.

I think a big point of this discussion was just trying to at least get manufacturers to word things in a straightforward way. Rather than just throwing big numbers around.

2 Likes
#23

I think we agree.

Funny enough though, this sounds silly to me:

If they use dishonest ratings today, how do you trust them in the future?

I think the only way to get real numbers is to have all tested with the exact same (physically same) setup by someone who has no interest in exaggerating numbers.

3 Likes
#24

Another thing to take into account is the massive swings in quality control that we have seen from certain ESC’s (cough flipsky cough). While it would still be good to have a test done so that the data is out there, one test done on one unit may or may not be indicative of what the average unit can handle. Im not really sure how to control for that, other than a larger testing sample size.

2 Likes
#25

Really easy way to control that.

#fuckflipsky

Just don’t buy it :man_shrugging:

7 Likes
#26

A trusted unbiased source that tests all escs in the same manner would be very welcome… we kinda do that already. I mean, I usually wait for one or two people i trust to buy a d review before i pull the trigger…

1 Like