Flipsky 75100 issue mitigation 1

[@o.m.2188]

Good job! Efficiency of the whole Controller should be much better now! Surface area of the housing is still the same but heat transfer and the ability to store heat temporarily is improved by almost 50%! For best heat transfer you should use Kapton MT tape in 0,2mm with TIM already applied on one side. It's more heat resistent, stronger and has better conductivity compared to normal Kapton tape (0,46 vs 0,23Wmk) and has pre-applied TIM especially made for such kind of tape (it sicks to the tape not to the mosfets because its silicone based TIM). This controller would be so nice to use it on ebikes because of its small housing; unfortunately there is no way to add a PAS sensor. I know, PAS integration is already realized in VESC 5.3 and above but most PAS Sensors available are not producing ADC signals. They only use hall sensors and 1-4v signals and we would need some kind of converter to get the needed ADC singal. I've not seen such a converter until now so current integration is useless for most of us. That a shame considering how powerful this tiny thing is.

[@Qdoggus]

Good job! I'll get mine later this month. It seems that you and I are working toward the same goals, and where we differ is in the connectors. I'll be doing xt90 for the power input, and MT60 for the phase output. I've used bullet connectors in the past with various RC/drone projects and never had an issue other than heatshrink rubbing thin and making contact. that was easy enough to solve though.

[@de-bodgery]

MT60 is super convenient since all 3 phases are in a single connector. I'm not overly fond of them because of the limited current handling...aka 60 amps. I do wish there was an "MT150" or something like the MT60 that used 5.5mm or 6mm bullets in it. I'd switch to it immediately! As is with the significantly improved current handling ability of the 5.5mm bullet, it's the best option in this amperage range. I'd rather have a little too much than not meet the maximums like the XT90 and MT60 connectors. Every time you have something limiting current, you are creating a place for losses and heat build up. It's the same problem as I just took care of inside the controllers. Why I'd want to add back losses with under powered connectors is not in my game plan. If your current requirements fit within the limits of your connector choices, it's not a problem. I don't know exactly how much current demand I'll have yet so I just went with "more than enough".

[@Qdoggus]

@@de-bodgery Excellent points, however, correct me if I'm wrong, the MT60 connector is rated for continuous 30A per bullet plug, 90A total split between the pulses of 3 phases. As for the XT90S, again correct me if I'm wrong but I read that it readily handles 90A sustained with 120A peak for 15 seconds.

[@Qdoggus]

@@de-bodgery Also, I use 175A sustained/250A peak Anderson style connectors with 8ga welding wire from the pack on the scooter, so you are definitely preaching to the choir.

[@de-bodgery]

@@Qdoggus MT60 is rated for 60 amps continuous for each 3.5mm bullet in the connector. If you had all 3 bullets in the MT60 in parallel all at the same time, then you could pull something like 180 amps. I'd say that 60 amps is pushing things for such a small bullet! Resistance and heat build up from such a small surface area for contact is going to create losses that are easy to avoid with a larger bullet. *** This is true for any motor and controller: At any one moment in time, you have current flowing into one phase wire and out another one. Both phase wires are "seeing" the exact same amount of current. 60 amps into one wire is still 60 amps leaving another wire. The connectors tying the motor wires to the controller will also see that same current in and current out. This isn't additive like if they were in parallel. *** XT90 bullets are 4.5mm and I think that's rated for 90 amps continuous...so I think you are correct about that. Still...If I'm looking for just barely able to do something, then any stress on the system will cause breakage, failure, losses and heat build-up at the weakest link. I prefer a margin of extra instead of "barely good enough". This is why I did the changes to my 75100 controllers...to eliminate the weak links and to give me a margin of extra.

[@Qdoggus]

@@de-bodgery Oh, I absolutely agree, "better to over-engineer than to under-engineer". My experience has led me to personally have to throw all the stated ratings out the window more times than I can count, at least for drones, scaled RC, and robotics. I've had xt60 connectors fail at 15A, and Deans connectors take way more than rated, for years. I suppose the most reliable way to assess capabilities are with calcs and actual torture testing. I 100% see the reasoning and logic in your choices, and one thing I forgot to mention was my intent to use software to detune the controller to 48v/35A for my purposes. If 4.5mm bullet connectors had more methods of retention, I'd definitely use them. At this point I'm planning and designing my own bus-bar distribution box that will definitely use copper (upper and lower bars) and ABS (outer shell) to bridge connections between phase input and output. Like you said, better to have excessive safety margins. I honestly find it a little strange that these smaller controllers don't have terminal points like Sabvoton does with the 72150.

[@Maker1000]

Nice mods ! Will be using these ideas for my 75100 which goes on my ebike.

[@DigBipper188]

Here's an idea; What about gluing the controller shells to a water block, then liquid cool the controllers as a single package?... A single 80mm or 120mm radiator should be more than adequate for dissipating all the heat those controllers output.

[@de-bodgery]

Yeah...why not? Won't work on my City since I don't have space for the water cooling. Also my controllers are mounted to the swing arm. I have a large hunk of aluminum to dump heat into. For some other implementation this is certainly an option.

[@Camilobakes]

Hey I was reading the article you did or are doing on the dualtron city. It seems very poorly built. I'm ignorant to the EV stuff, but very interested. My dualtron thunder goes gen controller went out and I could only imagine they are built the same. I bought a replacement for it. I bought the scooter in May of 2020 and has 900 miles on it and thought it was early for the front wheel controller going out. I was going to ask you what you thought might have happened. The only thing I can of is the day it went dead I was switching from two wheel to one wheel while I was hitting the throttle which I usually don't do I usually stop and then hit the switch. I thought you mentioned that the way they changed power modes is not the best method but was curious on your input..

[@de-bodgery]

It's not a steaming turd. The frame and machining is plenty good. It's the details that are not done well. How the lights or whatever are implemented needed more work to make them good. It needs more power to be sure, but it's not garbage or poorly made, just needs more refinement. *** I'm dropping off of DIY Electric Car. I've already pulled down all my pics and won't be back. They banned me for the actions of an arrogant asshole! Enough is enough! AS soon as they unban me, I'll be taking down my content 100% and then never returning. I don't have to deal with that garbage!

[@Zomby_Woof]

So what are you using these for? I've waffling between a pair of these, and a dual controller for my dual motor fat bike.

[@de-bodgery]

I'm using them on a stand-up scooter from Dualtron. It has 2 hub motors...so I need 2 controllers. What voltage are you running at? Flipsky has 2 controllers both named 75100 that are both 84v controllers. If you need 20S, I don't know of a dual controller that supports 100v max or 20S continuous. I think you are stuck with individual controllers if this is your pack voltage. Also, your phase wires are going to be pretty long to one of the 2 motors and this isn't a great idea. You would do much better to keep your motor wires short and have longer battery wires. Dual controllers would make this possible.

[@minalobudeshte]

how can we increase the initial torque on these controllers with the software? I tried changing the gear ratio but it doesn't have any effect. Any further ideas?

[@de-bodgery]

Go here...there is an entire VESC forum there: discord.gg/high-voltage-light-electric-vehicles-713249006129512448

[@ovidijuscybulicius4560]

3 weeks have passed already, what are the results?

[@sysghost]

My take on the whole waterproofing: I build my own scooters based on scrapped scooters complemented with regular bike parts. The compartments for the battery and the compartment for electronics (I keep them separate) are both waterproof rated to IP68. The only thing bringing the IP68 rating down a tad would be the cables going in/out of these compartments. (Marine/submarine quality cablings are expensive a.f.) Since everything inside never experience water or even moist, the connectors don't need to be IPx5 rated. I guess you got connectors and cabling exposed to the weather in such a way you need to do this.

[@de-bodgery]

correct

[@eelcohoogendoorn8044]

Any reason not to just dunk urethane down the whole enclosure after the electrical issues have been fixed? Just the weight difference?

[@de-bodgery]

@@eelcohoogendoorn8044 That would probably waterproof it. Thing is urethane is more of an insulator than a conductor of heat. You'll be trapping the large caps in their own heat. Air is a better conductor. I'd get thermal potting to fill the shells. That way you improve thermal transfer inside the shell and you water proof at the same time.

[@eelcohoogendoorn8044]

​@@de-bodgerywell doesnt have to be urethane; I guess whatever appropriate potting compound is available. But id guesstimate each cap has a max of about 1W of heat to diffuse; I think about any solid polymer should be able to handle that over the big cap area despite their low intrinsic conductivity.

[@eelcohoogendoorn8044]

Or maybe not I havnt ran the numbers on it...

[@lbabacar8653]

Hey man great job, do you have some clear picture ? I had v2 with 3 stacked shunt, i would rework a bit with copper shunt and add some copper, might do an heatsink again on the mill. Do you have a contact we could talk togheter and share me some of your picture

[@de-bodgery]

Getting my camera to stay focused was a PITA. I did 6 takes of this video and each time it would focus on the desk or something not the controller board. This was the best take...so I went with it. I have a build thread on DIY electric car and there it's mostly photos with links back to these videos. Set up an account on there and we can PM. I'm thinking about milling a heat sink as well or maybe just buy one and machine it to fit the controllers. *** Here's my build thread: www.diyelectriccar.com/threads/dualtron-city-upgrade.206509/

[@softart5235]

Urgent! could you build more mosfet radiators? Also I have lights and horn,Is there a way to connect those also ?.

[@ВладимирЛ-м7д]

Hello. Can you tell me what mosfets are installed in these controllers? What are the inscriptions on the mosfets?

[@de-bodgery]

Go watch my teardown videos. It's in there.

[@lusher00]

I have 2 of these I can't get to work. Is it possible I bricked them with the vesc tool by allow them to try a firmware update? It's been months since I picked these up but when I did I was just clicking buttons. I have the tools to reprogram the processor but I don't know where to find the image.

[@de-bodgery]

Yes it's possible, but unlikely, VESC is very resilient against bricking as a general statement. It has 2 storage locations in flash for firmware. It loads from one and then installs new firmware in the other. If something goes wrong, the CPU can still boot from the other storage location. This is never 100% sure, but it does add a layer of redundancy that most other firmware operations in other controllers lack! If the new firmware loads and boots successfully, only then does the backup older copy get over written with the new firmware. Having said that, you still have USB, UART and SWDIO for reflashing and recovery if they are bricked. Id try USB first...see if the PC based app can connect to them and then use VESCtool to update the firmware according to the video that Flipsky has on their channel. I've used that exact procedure many times on their controllers without issues. IF you need the firmware to use with an STlink and SWDIO, contact Flipsky to get that.

[@lusher00]

Well the primary symptom I’ve identified is no USB. I’ve powered it with a lab supply and I get a blue light but nothing enumerates when I plug in USB. I’ve tried multiple boards, multiple cables and multiple computers. I hooked everything up and applied a signal through the servo input just to see if it would respond but the unit isn’t even drawing enough current to boot the processor.

[@de-bodgery]

@@lusher00 Ping me on facebook. USB applies 5v to the controller and then the 3.3v supply ought to power up to and that's what powers the CPU. At least LED's ought to light up! Seems you have something else going on possibly and not bricked CPU's from a bad flash.

[@lusher00]

@@de-bodgery I’m not sure how to ping you on Facebook. I’ll have to investigate. I don’t believe Vbus is connected on the supplied USB cables. They are only 3 wire so I would assume D+/- & ground. I have LED’s on when I supply power but I don’t know what a solid blue LED means.

[@de-bodgery]

@@lusher00 Richard Gumfaw

[@gim8756]

hey! nice tutorial, i somehow need ur help, my controller is somehow damaged, when i connect it to any battery it starts to melt the connectors and get very hot, the light on the pcb stay one second then it goes off, is there any way i can repair it? mistakely i touched the phase wires that go to the motor, nothing is burned on the controller i really don't know what to do

[@de-bodgery]

You fried that one. Most likely you blew mosfets when you crossed the phase wires. These controllers are pretty cheap. Just get another one. It's possible replacing the mosfets will get it working again, but they also lack isolation between the high voltage side and the low voltage side. It's very possible that blowing the mosfets also damaged other things like the MCU.

[@de-bodgery]

Don't get another 75100 from Flipsky. There are other options that cost more, but much better built. The 75100 isn't a very good design as you can see in my videos. Get one of these options instead: flipsky.net/collections/electronic-products/products/flipsky-75100-with-aluminum-case-based-on-vesc-for-electric-skateboard-scooter-ebike-speed-controller or www.makerspev.com/products/little-focer-rev3-standard or spintend.com/collections/diy-electric-scooter-parts/products/single-ubox-100v-100a-motor-controller-based-on-vesc?variant=41797648875672

[@BLWN402]

@@de-bodgeryI was hoping you could help me out with VESC selection. I'm looking to replace the $6 controller in my kids ride on car but don't wanna spend $300-$400. It has a single 180w brushless motor... I've seen someone run this motor with a voltage of 75V/150A using 2-3x 6s LIpo 14AH batteries on a Flipsky VESC 6.6. I don't think I will need a 3rd battery to make it 18s since the speed is already enough for my kids, maybe just for the better range. My problem is this car has a VERY hard time with any hills or inclines so that's the main reason for the upgrade.. I'm looking for a VESC that I can run around 120A max for motor (60A continuous and 60v). I would like to test it on the 24v/14Ah/20Hr Lead acid battery now if possible before an upgrade to a 2x 6s Lipo (12s - 48v) setup. Any help you can provide would be greatly appreciated! Here is my car: www.bigtoysgreencountry.com/fast-24v-spartan-big-kids-ride-on-buggy-xxl-180w-motor-air-filled-rubber-tires-red/ (180W Brushless 24V motor)

[@de-bodgery]

@@BLWN402 1. It is unlikely your SLA's will do a very good job. They will seriously lack the current delivery you are talking about and simply sag BADLY! So each one delivers 12v or so under no load. Put them under 120 amps and you'll something like 6v or maybe 7v from them and maybe 1/3rd of that amperage. Basicly SLA's suck really badly! Expect to be replacing them immediately. 2. 60v VESC is pretty common, but if you ever want to upgrade it later for more speed, you might as well go with 75v or 100v controllers instead. The cost difference is small. You might also want to pull that VESC controller off and use it elsewhere later too. 75v or 100v controllers gives you more flexibility for whatever might happen later. The higher voltage versions run just fine at lower voltages so this is not a problem. 3. Your 180w motor. It's probably an inrunner? Can you point me to some place online where people are using it at 150a and 75v? I'm much less concerned about the voltage than the amperage. And I have to say 11,250 watts from a motor meant for use at 180 watts is pretty much impossible! I over watt motors all the time and push them right to their limits and run them there long term. I never NOT EVER see 62X MORE wattage from a motor! I'm highly skeptical that those are real numbers! Go look at my Zap scooter videos. That's definitely the BEST over watting I've ever done and it's still just about 7X more wattage. I can push that motor harder, but it gets really hot and I'm already seeing peaks of 140C in it. I REALLY don't want it to EVER get hotter than 150C as then it will burn out! 62X more wattage...this is highly improbable! *** So then VESC options that are probably overkill for your motor, but meet or exceed what you talked about: (There's more options than these) 1. flipsky.net/collections/electronic-products/products/flipsky-75100-with-aluminum-case-based-on-vesc-for-electric-skateboard-scooter-ebike-speed-controller 2. flipsky.net/collections/flash-deals/products/flash-deals-flipsky-75200-84v-high-current-with-aluminum-pcb-based-on-vesc-for-fighting-robot-surfboard-agv-robot 3. www.makerx-tech.com/products/makerx-go-foc-hi100-75v-100a-base-on-75300-hardware-pre-sale 4. trampaboards.com/vesc-6-75-complete--perfect-performance-running-up-to-16s-po-p-33317.html 5. spintend.com/collections/diy-electric-scooter-parts/products/single-ubox-100v-100a-motor-controller-based-on-vesc 6. spintend.com/collections/diy-electric-scooter-parts/products/75v-100a-single-esc-single-ubox-based-on-vesc 7. maytech.cn/collections/all-speed-controllers/products/mtvesc7-12-200a-50v-vesc

[@BLWN402]

@@de-bodgery Thanks for the info... Those are pretty much all of the VESCs I have on my list that I'm trying to decide from. I'd like to find one available in the US if possible. Would your #1 Flipsky 75100 be my best choice for the money? Most of the other ones are not available right now or ship from China unless you know of a US based supplier for any of them. I will do my battery upgrade at the same time I suppose. I'll plan on 2x 6S batteries and run it at 12S/48V. This video here is a guy that's made a bunch of upgrades to his car that has the same 180w Brushless motor/differential that I have. I figured with a 12S battery and 60amps I'd be fine as he's ran his way higher on an 18s battery. Thoughts? Thanks!!! kzbin.info/www/bejne/qYnWe3SHrbV2pbs

[@mahmutkurt9884]

Hi. what is the moss number for this?

[@de-bodgery]

did you watch the videos I did on these controllers? Also, moss is green and tends to grow on logs or trees. They are mosfets, not moss...not even spelled the same. AND within 7 seconds of this video starting you can see for yourself exactly how many mosfets are in the controllers.

[@mahmutkurt9884]

@@de-bodgery sory for thats, mossfet. only want to learn what is mossfet number. ırfz44n like thats.

[@SiFFon-h2j]

SHIT !!

[@helios_ja]

The people asked and De-bodgery answered! Thank you so much for this video kind sir! This is going to be the best entry level controller on the market until ftex.ca/product/ganrunner-road/ are available. I look foward earnestly to your teardown and inspection of it once it gets to market. I hope by then, I'll be able to finance some of the cost towards the purchase of it just to show my appreciation!

[@de-bodgery]

Galium nitride mosfets are slightly better than the stuff available now in most controllers. You'll see them get adopted as the price for them comes down and options in lots of voltage and current ranges become more available. For now at least, current mosfets are pretty rock solid and perform well. GAN isn't a game changer, just an incremental improvement....at least for now. Maybe in a few more years we'll see GAN make current mosfet technology obsolete, but I doubt it. This is a change like LION cells from 5 years ago vs modern LION cells. *** They are cheap...no doubts there! I do hope that Flipsky takes everything I show here and in my teardown video and incorporates them in a new revision of the 75100. It's sad they didn't bother making these things as good as their other products!

[@de-bodgery]

I hadn't spent any time looking at GAN mosfets...so I spent a few minutes looking at what Digikey had. I'm not overly impressed! Most are low amperage and can't even compete with "standard" mosfets for current handling. This chemistry needs lots more work before it can displace current mosfet technology!