Did you cut off the 5V on one of the ppm cable?
Yes i have
No time so far to ride again. Hopefully some time over the weekend weather permitting.
@Benjamin899 … we did … see previous post. It was the second one we tested. Worked OK, but case is not a heat sink so max watts constant are same as BV video on v4.12 hardware … approx 23v / 34a constant. No particular issues with them. Didn’t fail during tests. The are spending a lot of time getting this right. Once we’ve tested the V6 with heatsink case, I’ll share the data. Cheers
Dude, that’s dangerous, the full current of the FSESC that you didn’t turn on is running THROUGH YOUR SPLIT PPM WIRE.
The on/off switch is a low side switch, the split ppm wire connects the grounds of the two FSESCs together. When one switches on, the other one now has a path to ground though the PPM wire.
Disconnect one of the ground wires on your split PPM cable.
MAKE SURE YOU SWITCH BOTH OF THEM ON OTHERWISE YOU’LL BLOW UP YOUR PPM CABLE.
This is why you don’t f***ing implement a low side switch for this sort of application.
IMO, looking at the board they had plenty of space to add something like an LTC7000 to do high side switching with the 3 N-channel MOSFETs.
Alright, so I went to the hilliest place I know near me… This was with a dual hubmotor build on FOC.
These thing help up very strong!
Did go from 100-70% in about 3miles in 20minutes so range went down a bit given the terrain.
Temp went from 100 F to 115 F… It was about 105 F outside…
This isn’t my normal test ride since I don’t really go there too often. So this test is pretty limited, I can’t exactly put hundreds of miles in this place in a reasonable amount of time. However from what I experienced today I believe these ESCs are very good right now.
I’ll keep them on FOC and use as my main ESCs for right now.
I disconnected one of the positive wires on my PPM cable. That not correct?
No, you must disconnect the negative wire on one of them.
The negative wire is what connects the grounds of the ESCs together.
If you leave both of them attached then you must make sure both ESCs are physically switched on before you use them. If only one is switched on, then the other “thinks” it is on because it sees ground through the negative wire of the PPM cable.
What the on/off switches on these ESCs do is it connects and disconnects the ESC from the negative terminal on the XT90 connector. What your PPM cable is essentially doing is bypassing this through the negative wires.
I run only the PPM wire to the second vesc, it works for me.
ahhhh… ok now it makes sense.
What I want in the end is to have both ESC’s connected via split PPM but if the ESC that is supplying power to the recevier dies the other ESC will be able to power it. I’m going to assume that a BEC is what I’m going to need?
There are two ways you can do this, BEC or (Ideal) ORing Diodes.
A BEC hooked up to your battery will do the job just fine. Connecting this to your receiver means you only need to connect the signal wires to the ESCs. It’ll be the simplest and easiest solution so I recommend this.
If you don’t want to get a BEC, the other solution is a little more tricky.
You need either a pair of Ideal Diode ICs or
a pair of Shottky Diodes with a very low forward voltage drop.
Normal diodes would be much simpler and easier to wire up
Something like this
looks like it will do the job. (Forward drop of ~300mV at 1A).
What you need to do is splice the POSITIVE 5V wires and solder these shottky diodes IN SERIES, with the cathode POINTING AWAY from the ESC. So this means each ESC will have a shottky diode pointing away from it on the 5V wire, before they both meet at the PPM splitter.
It will look something like this
ESC#1 +5V -----------|>|-----------Splitter
ESC#2 +5V -----------|>|-----------Splitter
Both options are perfectly safe done correctly. (The latter is employed in automotive and industrial applications. It is called Redundant Power Supplies.)
Honestly, it is a huge oversight that the VESC and its derivatives don’t already employ ORing diodes on their 5V outputs when it was expected that many people would be using more than one.
Grr, makes me want to develop my own ESC that fixes all the hardware problems/oversights present on the VESC and its derivatives.
That is where this forum is split. Those who are EE majors and those who are educated hobbyiest. I haven’t heard of an ORing diode configuration until you said so. I tried on mine to add a TVS diode to places not original in the VESC 6.4 schematic. It’s probably two more layout changes before I finally call my derivative done.
Really? (Ideal) ORing diodes are pretty standard practice in applications with multiple power supplies. I’d be really surprised if people here aren’t already doing this.
I’m not an EE major, I just have lots of time and spend it all reading application notes haha.
Also, I realized my previous post was rather ambiguous so I updated it with a diagram and better description.
Won’t I need to do this to the negative so it does not bypass the anti spark?
Forgive my ignorance, but can someone tell me about the built in anti spark. If you leave the switch off, can you just connect your Vedder switch in place of the Flipsky built in AS. Or can you leave it on, and still connect the vedder switch with your 60amp fuse to this ESC?
You can leave it on and use the vedder switch and fuse to turn on and off.
It is basically having 2 antisparks in series. Both have to be on for the ESC to turn on
Thanks for the response
So of the three wires comming from the ppm cable, is the servo wire comming from the MCU or the 5V rail more likely to be affected by a random transient? There was a warning posted about using Y-splitter for the ppm and resulting in destruction of the MCU but I could not figue out which line to put the diode on on my version.
Yes, each ESC needs a diode pointing towards it on the negative line.
It will look something like this
ESC#1 GND ------------|<|---------Splitter
ESC#2 GND ------------|<|---------Splitter
You may get a bit of distortion on your PPM signal, so be sure to tune their input configurations accordingly.
I would think the PPM signal is more susceptible to transients, but there is a resistor in series with the signal and the GPIO on the STM32F405RG, which (in combination with the GPIO’s clamping diodes) should protect it from EMI or transients generated by the receiver. Or if you accidentally configure it as an output, the resistor should limit the current flowing in and out of the GPIO. Is it the MCU that explodes or the DRV8302 that explodes? Because only the signal wire is connected to the MCU, +5V comes off the DRV8302, and GND is connected to Digital GND of the ESC.
I can see why the DRV8302 might explode if the 5V lines are connected together WITHOUT ORing diodes, because if one of them goes into fault, it’ll start sinking current from the other DRV8302’s buck converter and cause both of them to fail.
I’m pretty sure you need to put the diode here.
Between the output capacitor and the feedback resistors, the cathode of the diode goes to +5V.
However, if your VESC variant does not have those resistors in place, then it is the signal wires that causes the MCU to blow up.
Now that think more about it, even with the diodes in place, things might still go wrong and the STM32F405RG might blow up through the signal wires.
I would recommend turning both ESCs on before plugging in any shared wires.
I would feel really bad if the $10 MCU blew up on one of your ESCs.