10S custom skate ESC: testers wanted!

Hi

I finished building 2 esc this week. this give me the possibility to setup some kind of bench test with 2 motors

power regen
batt >esc >motor-motor<esc<batt

With this can run High continous Amperes.
motors are sk3 6374-149kv
I am able to run 50A input for 30s before fet are at 75C no cooling (clamp meter)
there is a notable difference between esc measured current and clamp meter.
I dont know which to beleive.
Overall efficiency esc+motor seem to be 83% but that really rough measurements.
Motors seem to handle that pretty well but will overheat without any cooling.

for example i get 19A regen with esc2 + 9A batt that gives 28A into the motor esc1
I dont know how well my esc are built this is my first electronic project. based on this I would recommend 25A continous
This is 1000W on 12S wich is not bad.

I wil setup cooling and get back with results.

https://drive.google.com/file/d/0B7HfQHbqKf_lcnllUEE2WGFkeGRKOTRJbnZrZE5WdTAyRGhZ/view?usp=sharing
Hope this help
 
Joe90 said:
Hi

I finished building 2 esc this week. this give me the possibility to setup some kind of bench test with 2 motors

power regen
batt >esc >motor-motor<esc<batt

With this can run High continous Amperes.
motors are sk3 6374-149kv
I am able to run 50A input for 30s before fet are at 75C no cooling (clamp meter)
there is a notable difference between esc measured current and clamp meter.
I dont know which to beleive.
Overall efficiency esc+motor seem to be 83% but that really rough measurements.
Motors seem to handle that pretty well but will overheat without any cooling.

for example i get 19A regen with esc2 + 9A batt that gives 28A into the motor esc1
I dont know how well my esc are built this is my first electronic project. based on this I would recommend 25A continous
This is 1000W on 12S wich is not bad.

I wil setup cooling and get back with results.

Hope this help
Click to expand...

The thermistor chip is a few mm away from the fet's so it will not read an exact temp of the fets just the board temp in that area. Also since it is on the side of the board you will not get the temp you will get from the middle of the board/fets which will be more hot. Are you using the newer IRFS7530TRL7PP mosfets or the older mosfets?
 
Fully Assembled with 5.5mm bullets & Cap installed..... also servo wires so you can plug your RX straight in... Yes we also hate soldering as much as you do!!

in regards to shipping schedule... it seems we are currently on track!.

vesc.jpg
 
I use IRFS7530

Just gentle blowing from my mouth make the temperature drop. I am confident that with a fan and heat sink we can pull much more
 
I think, I ll try to put the FETs in a sndwitch of aluminium plates connected to a heatwink.
Not sure the fan is needed if the sink is put in the board airflow.
 
Joe90 said:
Hi

I finished building 2 esc this week. this give me the possibility to setup some kind of bench test with 2 motors

power regen
batt >esc >motor-motor<esc<batt

With this can run High continous Amperes.
motors are sk3 6374-149kv
I am able to run 50A input for 30s before fet are at 75C no cooling (clamp meter)
there is a notable difference between esc measured current and clamp meter.
I dont know which to beleive.
Overall efficiency esc+motor seem to be 83% but that really rough measurements.
Motors seem to handle that pretty well but will overheat without any cooling.

for example i get 19A regen with esc2 + 9A batt that gives 28A into the motor esc1
I dont know how well my esc are built this is my first electronic project. based on this I would recommend 25A continous
This is 1000W on 12S wich is not bad.

I wil setup cooling and get back with results.

https://drive.google.com/file/d/0B7HfQHbqKf_lcnllUEE2WGFkeGRKOTRJbnZrZE5WdTAyRGhZ/view?usp=sharing
Hope this help
Click to expand...

Nice that you got it working and made a test like that.

What is causing heat is the motor current and not the battery current. You can run 10A battery current and very low duty cycle with high torque and still heat up the FETs without drawing much from the battery. Because of this, it is important to get the motor kv, gearing and battery voltage right. Also, it is important to notice that just reducing current a little bit will cause significantly less heat because the heating is proportional to the square of the current.

In your test, to get the highest battery current, you should set the driving ESC to duty cycle mode and run 100% duty cycle, while running the braking ESC in current control mode with different braking currents. Use a low voltage battery, like 3s, to not have too high speeds on the motors to prevent hurting yourself.

Regarding cooling, just a little bit of airflow make a huge difference.

Regarding the temperature limit, it should be quite safe to leave it to the ESCs to limit output power when the FETs are getting too hot. I have hit the temperature limit several times with my single-motor 8s board with a current limit of 80A.

The thermistor chip is a few mm away from the fet's so it will not read an exact temp of the fets just the board temp in that area. Also since it is on the side of the board you will not get the temp you will get from the middle of the board/fets which will be more hot. Are you using the newer IRFS7530TRL7PP mosfets or the older mosfets?
Click to expand...

The board conducts heat very well, so the thermistor reading is quite OK. I tested that with an IR camera that I borrowed once, and got within 10 degC agreement.
 
vedder said:
The thermistor chip is a few mm away from the fet's so it will not read an exact temp of the fets just the board temp in that area. Also since it is on the side of the board you will not get the temp you will get from the middle of the board/fets which will be more hot. Are you using the newer IRFS7530TRL7PP mosfets or the older mosfets?
Click to expand...

The board conducts heat very well, so the thermistor reading is quite OK. I tested that with an IR camera that I borrowed once, and got within 10 degC agreement.
Click to expand...

If I removed the thermistor and soldered tiny wires to it and the board then extended it to the top of the mosfet would it still work? Maybe get an exact reading? I will not even be concerned with overheating the mosfets, I would assume even 100c or more continuous would be fine for them based on the specifications they state. I know from repairing computer motherboards that continuous high heat will damage circuit boards though. My plan is to have the entire board be the same temp by adding thermal pads inside the copper case I am making, as the fet's warm up the entire circuit board warms up so there will be no thermal shock.
 
Joe90 said:
I use IRFS7530

Just gentle blowing from my mouth make the temperature drop. I am confident that with a fan and heat sink we can pull much more
Click to expand...

Why did you use those? I see they are the same price as IRFS7530TRL7PP. That thermistor will cool itself quick if it gets air flow and the reading will not reflect the board temp.
 
Sorry for not posting the complete part number, I use those specified by Benjamin Vedder
Mouser part : 942-IRFS7530TRL7PP

M. Vedder : thanks for your insight.
 
Joe90 said:
Sorry for not posting the complete part number, I use those specified by Benjamin Vedder
Mouser part : 942-IRFS7530TRL7PP

M. Vedder : thanks for your insight.
Click to expand...

ah, ok. Still waiting on parts to finish my scooter before I can even test mine. Are you able to get yours on a bike, skateboard....?
 
I did some more experiment bench running with my motor/generator

My phase current was practically equal to battery current. The duty cycle was 0.95
I tried 1.0 duty but it made no discernable change
Tried 0.50 and yes the battery current ended up = 0.5 x phase current.

What is the purpose of the ''spikes'' in phase voltage when loaded at 0.50 duty ?
Also the generator motor heat less, Is it the nature of the thing or energy recovery could be improved ?
The generator seem not able to load the motor more that 55A, is this because it's the same KV on the same voltage ?
And when I want to load it to the max, the regen part let go. but get back as I lower it to 90% regen capacity wich is more amperes that batteries can usually take.

I will put them inside a forced ventilation enclosure and see what happen.
I would be happy with 100A for 5 seconds. and 25A continous

Regards
 
Joe90 said:
I did some more experiment bench running with my motor/generator

My phase current was practically equal to battery current. The duty cycle was 0.95
I tried 1.0 duty but it made no discernable change
Tried 0.50 and yes the battery current ended up = 0.5 x phase current.

What is the purpose of the ''spikes'' in phase voltage when loaded at 0.50 duty ?
Also the generator motor heat less, Is it the nature of the thing or energy recovery could be improved ?
The generator seem not able to load the motor more that 55A, is this because it's the same KV on the same voltage ?
And when I want to load it to the max, the regen part let go. but get back as I lower it to 90% regen capacity wich is more amperes that batteries can usually take.

I will put them inside a forced ventilation enclosure and see what happen.
I would be happy with 100A for 5 seconds. and 25A continous

Regards
Click to expand...

The spikes in the phase voltage under load are caused by the inductance of the motor, when the coils try to resist a fast rate of change in current when a phase is disconnected during a commutation step. This is normal, but a bit tricky to deal with in software.

The reason that you cannot brake with more than 55A (if you have set higher current limits than so) is probably that VESC cannot do active braking (active braking = putting power into the motor to slow it down, giving negative efficiency), only regenerative braking (doing active braking is very tricky). To get more braking current, you have to increase the voltage of your battery a bit.

VESC should be able to do 100A for 5s and 25A continuously without extra cooling, depending on how frequent the 100A loads come. On my 8s single-motor longboard I have set the current limit to 80A, and doing a full acceleration from standstill to full speed (about 37 km/h) is no problem without any cooling. The temperature limit does kick in when I go up very long and steep hills though. The same longboard with 12s and a lower-kv motor and otherwise the same specs would not have any problems with the hills.

/Benjamin
 
Is there anyone selling the latest version of this ESC right now or in the near future? (Preferably in Europe for ease of shipping)
 
last one took me 9 hours to solder
+ 3 toubleshooting

But you forget all that when the motor spin !

Most of my problems were with fet driver.
Benjamin recommandations to solder the driver are the best I think.
Unfortunately I read them after I had finished soldering both and tried differents techniques.
I was afraid to use the pen to solder such tiny leg but it's not so hard.
I used hot air for 90% off the components. It works pretty well.

If you like learning, you will enjoy it.
 
Joe90 said:
last one took me 9 hours to solder
+ 3 toubleshooting

But you forget all that when the motor spin !

Most of my problems were with fet driver.
Benjamin recommandations to solder the driver are the best I think.
Unfortunately I read them after I had finished soldering both and tried differents techniques.
I was afraid to use the pen to solder such tiny leg but it's not so hard.
I used hot air for 90% off the components. It works pretty well.

If you like learning, you will enjoy it.
Click to expand...

I did my first one by hand only and got a bad cramp. Now I do the entire thing other then the pinned chips and larger componets with hot air and board heat, takes longer but gets it done right and without pain.
 
You must log in or register to reply here.

Similar threads

T
Wireless Wii Nunchuck *TESTERS WANTED
14 15 16 17 18
Replies
440
Views
64,096
austindavid
A
B
DUAL E-MTB BUILD
2 3
Replies
68
Views
9,308
torqueboards
T
T
FREE Step By Step 6S Electric Longboard Build Guide
2
Replies
42
Views
54,847
torqueboards
T
D
How to build an Electric skateboard tutorial *free version*
2 3 4
Replies
82
Views
119,960
SirDuckington
S
Reid Welch
"Extreme MY1018" archival, closed out
Replies
11
Views
15,907
GIGATT
G
Back
Top