This is just a simple teardown and overview of the internals of the Enertion FOCBOX, its similar to my Shred Lights Review. Starting off I ordered mine during the Cyber Monday sale like a lot of people on 11-26 and received them on 12-23 so just under a month from China which is pretty good being it was with the free shipping. However the tracking number I received was either wrong or didnt work because no tracking web service could find it. Im sure I could have contacted Enertion about this but I thought I would just give it some more time until eventually a box showed up with a different tracking number and I had my FOCBOX’s. Overall the packaging was decent just some standard bubble wrap in a small box so no complaints there, pic below.
The actual ESC’s themselves look just beautiful compared to your normal bare board VESC with its plastic top and milled aluminum bottom. I also really appreciate the sticker labeling all the pins and LED’s which is a nice touch especially for beginners, and the MICRO USB port is a wonderful change which I welcomed with open arms compared to the standard mini usb trash, I do however wish the port protruded out farther as its relatively hard to plug some cables into it. The addition of the servo cable coming right out of the box for connecting straight to a receiver is another nice touch especially for beginners as well. The housing is held together with three M3x8 screws which go through the top plastic housing, pcb and into the metal bottom. The ESC is designed to have all the FET’s on the bottom side to dissipate all the heat through the metal heatsink and uses a piece of thermal tape in between. For the input it comes with a female XT60U-M which is just a slightly shorter version of your standard XT60 female plug and it works fine with normal XT60’s. The motor phase wires are 3.5mm bullets which like others have stated are a weird size as most motors are 4mm bullets, but I just ordered some 3.5mm male bullets and soldered adapters together for my 4mm motors. Making adapters is usually far easier than trying to solder bullets onto copper wire. The FOCBOX also weighs in at 105 grams, which I cant compare to anything else as I never weighed my VESC’s. On the bottom of the FOCBOX there are four M3x4 Mounting holes which are 30mm wide and 15mm tall which I suppose you could use for mounting to a larger heatsink or maybe an enclosure, idk.
Moving onto the best part, the actual internals and electronics inside. To start the FOCBOX has two 680uF 63V Nichicon Caps compared to for example my Maytech VESC which has three 680uF caps, but I think that still 1360uF is plenty. For the most part everything’s the same as the VESC 4.12 on the FOCBOX other than the shape and DirectFET’s but there are still a few smaller changes. For example the TVS diode seems to be different though I cant find any information on the one in the FOCBOX, its labeled as GFZ 74A though if anyone can find any info on it. Other than that all the other parts seem the same DRV8302, STM32F4 MCU, SN65HVD232 Can Transceiver, etc. The MOSFET’s are IRF7749’s with a max Vdss of 60V which works though I would have prefered to see maybe the IRF7759 though it has a higher RDSon of 1.8mOhms compared to the 1.1mOhms of the 7749 so I can see why they went with it. Another thing to point out is the silicone all around the capacitors and main power leads which I actually like a lot because it seems to keep everything together quite well. You might also notice the white gunk on the TVS diode near the MOSFET’s which its actually like on all my FOCBOX’s or at least to some extent.
Last but not least I heard some people talking about whether or not the FOCBOX has a conformal coating which I can confirm it does not. Its not a big deal as if water is getting into your enclosure you probably have bigger problems then messing up your ESC though it would have been a nice feature, nonetheless I added my own coating of MG Chemicals Acrylic Conformal Coating. First pics are the before with a blacklight and last pics are after the coating was applied.
Overall the FOCBOX is a great ESC and I cant wait to test it out on my board once it gets a little warmer out, its currently about 5°F in Chicago right now, but once I do ill either update this post or add a reply. FOCBOX Schematics can be found HERE on the Enertion website. Also if I missed anything or if there’s anything you’d like added for example a feature or part I missed about the FOCBOX please let me know and I will add it!
Acrylic Conformal coating on fets does not prevent heat dissipation. It’s also the only thing that can protect fets from water. Oil based sealants don’t work for fets.
Only use acrylic based conformal coating on fets, silicon conformal coating is to thick and will trap heat, and oil based(corrosionX) will not prevent water damage.
You’re lucky, I ordered a couple FocBoxs as well on Nov 28th and just got an email yesterday saying that they have to make more with a new manufacturer to fulfill all the backorders they have.
I don’t want to say that conformal coating (event the acrylic one) won’t work because it could work, but it could also reduce the thermal performance of the fet and heat sink. And since it hasn’t been tested by Enertion over their product, if something blow up it could be seen as a void of waranty.
They are talking about something different there.
The op on that link was asking about coating over the heatsink itself which is a bigger hindrance than between the heatsink and component.
I agree though, there are a few thermally conductive conformal coatings but from my research acrylic isn’t one of them. MG chemicals doesn’t list the thermal conductivity of their stuff on the tds and I assume it’s not very high.
I have never covered any component that is meant to dissipate heat.
Sure, it could void the warranty. But your also talking about a coating applied properly should be less then 0.15mm thick. Or 150microns. While the coating is low in heat conductivity, a few microns will be negligible when talking about percentages of a degree.
As well the manufacturer warranty is only good for 60 days, the purchasable year long warranty supports all user modifications with the exception of SMD component changes according to the wording.
It’s the same argument as electrically insulating your packs with kapton. Protection > negligible temp variance.
Hey @JohnnyMeduse@longhairedboy@PXSS I did cover the FET’s in a thin layer of the coating but I tried to keep it thin and only did one coat unlike how I do two coats on some areas, ie DRV and MCU. I have coated MOSFET’s and other components connected to heatsinks before with thermal compound in-between and I haven’t seen any noticeable difference. I will however test this out because it’s a good point so I’ll let everyone know with my findings! And the main reason I covered the top of the FET’S was for corrosion resistance because they are metal unlike normal D2PAK’s which are plastic.
You can find all the main differences on the thread by Onloop HERE. It states that the main differences are DirectFET package MOSFET’s which greatly improve thermal transfer and make it so the FOCBOX can handle higher currents than a VESC. Increased number of decoupling capacitors, which you can see on the top of the PCB near the shunts, to improve reliability in FOC I assume as well as a Micro USB port and some other changes. And I think maybe the biggest difference is the device itself, it really feels like an actual product and not some DIY raw PCB that has so many exposed pads. The metal and plastic case really gives it a lot of protection and like Enertion said makes it far more manufacturable than your normal VESC.
Alright guys so finally got around to testing the thermal conductivity of the Acrylic conformal coatings! Let me start by saying that a single layer of the coating does absolutely nothing to the thermal transfer. To test this I used a internally shorted IRLZ44N MOSFET attached to a small heatsink drawing 12.5W. I first tested without any coating at all just screwed together and recorded the temps at different time intervals of 15s, 30s 1m, 2m and 2m 30s. The second test was with a single normal thickness coat that was thick enough to electrically isolate(tested with multimeter probes). I was planning on doing 2 coats as well as 3 and so on until I started seeing a difference, until my MOSFET suddenly stopped being shorted so I need to find a different way of testing this. I plan to figure out a new method of testing tomorrow and start again. Anyways below is my test data, pics are of the simple setup and measurements where taken from the center of the back of the heatsink.
based on what i’m reading here this test proves that a thin coating limits the heat transfer by as much as a full 2 degrees C in some situations over the same time interval. So while the final temps are the same, the transfer rate is indeed throttled.
It does do more than exactly nothing. Little more, but more.
Well looking at the 0-15S interval the coated transfer is 4c higher, the 30-60s interval is also higher with the coating as well. This is why I’m going to redo all the tests again and do them multiple times because this premlinary test seems a little odd.
10char
You have the tests well laid out too!