full-throttle
1 MW
Bring it to work, we'll figure it outadrian_sm said:
'should' is a strange word..
-
Howdy! we're looking for donations to finish custom knowledgebase software for this forum. Please see our Funding drive thread
Commuter Booster - <1kg Friction Drive
- Thread starter adrian_sm
- Start date
adrian_sm
1 MW
Update: Road Bar Button Throttle
Made a better button throttle that I think will be great with the ramp control, and future power limiting.
View attachment 1
Made a better button throttle that I think will be great with the ramp control, and future power limiting.
View attachment 1
adrian_sm
1 MW
Update: ESC-BEC Light Power Adapter
This is before the final shrink wrap, but it allows me to power my magicshine light from the throttle 5v.
This is before the final shrink wrap, but it allows me to power my magicshine light from the throttle 5v.
adrian_sm
1 MW
Update: Battery options on Road Bike
Here is the road bike with the long range 20Ah, and normal 10Ah battery packs.
Here is the road bike with the long range 20Ah, and normal 10Ah battery packs.
Kepler
10 MW
Thats a nice button setup Adrian. Where did you get the button? I just sold my flatbar and bought a full carbon, full Shimano Ultegra equipped road bike and are looking at options in relation to fitting a button to the race style bars.
adrian_sm
1 MW
You sold the flat bar! That was a sweet little bike.
The button is from Jaycar. Designed for through hole mounting, and not water proof. But it is really nice, light touch so not hard to hold down for extended periods of time, low profile so much easier to mount. Definitely the best option I have found.
The button is from Jaycar. Designed for through hole mounting, and not water proof. But it is really nice, light touch so not hard to hold down for extended periods of time, low profile so much easier to mount. Definitely the best option I have found.
There are some smd resistor a 0 ohm conecting PA3-7 together thet i hade to remove to get in and output work.
They was conected to +3,3 what I remember in wattmeter .
If it all work you vill get /2 the input from the hall to the A/D and the referens voltage is set to 2.56V .
Vill be aprox 0-1024 bit if you get 0-5V in
//Jonas
They was conected to +3,3 what I remember in wattmeter .
If it all work you vill get /2 the input from the hall to the A/D and the referens voltage is set to 2.56V .
Vill be aprox 0-1024 bit if you get 0-5V in
//Jonas
adrian_sm
1 MW
Ah that explains it. Looking at the PCB images I can see that now.
As I haven't pulled the two PCB apart, it is going to be challenging to remove these resistors. Getting a soldering iron in there is very tight. I might just drill through the board to break the traces between the via holes and the resistors.
full-throttle also thinks a diode on the throttle signal , wil removing the Jonas's resistor might also do the trick. Time will tell.
Then I will need to look at the voltage range of the hall throttle, as I think it will only be ~1.0 - 4.0 V range. I'll have a closer look at this part of the code and see what I need to do.
As for the little programmer, it was just a dodgy connection as I was pushing an IDC header in to the holes on the PCB, and some of the pins were not making connection. Now that I have a right angle IDC header, with more flexible and longer legs, I am making good contact, and can now program. Woo hoo. Already tweaked a few things in the code, and reflashed successfully.
As I haven't pulled the two PCB apart, it is going to be challenging to remove these resistors. Getting a soldering iron in there is very tight. I might just drill through the board to break the traces between the via holes and the resistors.
full-throttle also thinks a diode on the throttle signal , wil removing the Jonas's resistor might also do the trick. Time will tell.
Then I will need to look at the voltage range of the hall throttle, as I think it will only be ~1.0 - 4.0 V range. I'll have a closer look at this part of the code and see what I need to do.
As for the little programmer, it was just a dodgy connection as I was pushing an IDC header in to the holes on the PCB, and some of the pins were not making connection. Now that I have a right angle IDC header, with more flexible and longer legs, I am making good contact, and can now program. Woo hoo. Already tweaked a few things in the code, and reflashed successfully.
adrian_sm
1 MW
Managed to desolder the 4 offendng SMD resistors. Just pried the boards apart a bit then jammed the solder iron tip in there. Done.
So I now have a more linear resposne for the current control pot, and the throttle signal is looking better. With the screen telling me I am going from ~1200us to 1920us. But obviously not going down to zero so the ESC isn't arming yet.
Now the bad news. I was interested to see what the throttle out signal looked like so I thought I would take advantage of the programmers ability to be a oscilloscope. So I connected connected the throttle signal to the programmers A IO, and then things went wrong. Programmer no longer works.
I was powering the watt meter and esc from a LiPo battery, and the programmer for the laptop on battery power.... I thought this would be okay, but apparently not. Guess I should have grounded things first....
Arrrrggghhhhhh.... Just when i get the hardware right, the programmer talking and can reflash the watt meter at will, then i go and screw it up..... Bugger.
So I now have a more linear resposne for the current control pot, and the throttle signal is looking better. With the screen telling me I am going from ~1200us to 1920us. But obviously not going down to zero so the ESC isn't arming yet.
Now the bad news. I was interested to see what the throttle out signal looked like so I thought I would take advantage of the programmers ability to be a oscilloscope. So I connected connected the throttle signal to the programmers A IO, and then things went wrong. Programmer no longer works.
I was powering the watt meter and esc from a LiPo battery, and the programmer for the laptop on battery power.... I thought this would be okay, but apparently not. Guess I should have grounded things first....
Arrrrggghhhhhh.... Just when i get the hardware right, the programmer talking and can reflash the watt meter at will, then i go and screw it up..... Bugger.
Any way Adrian if you get 1200 -> 1920 uS you can test seting up ESC to that range I was only learning ESC the new trottel range whene I have hall trottel.
the ESC I have will start at 14-1500uS but can be diffent for you.
The range is 1000uS to 2000uS if you have about 0-5V but 1-4V fill work to you only learn the ESC that.
Sorry your programer broke I hade also learn the hard way to not have it conected if you have more than 5V powering the wattmetter .
Can you see the output with some other oscilloscope ?
/Jonas
the ESC I have will start at 14-1500uS but can be diffent for you.
The range is 1000uS to 2000uS if you have about 0-5V but 1-4V fill work to you only learn the ESC that.
Sorry your programer broke I hade also learn the hard way to not have it conected if you have more than 5V powering the wattmetter .
Can you see the output with some other oscilloscope ?
/Jonas
adrian_sm
1 MW
umejopa said:
I thought so too. But The ESC would not learn, that was why I was trying to get the osciloscope on it.
I'll hopefully drag it in to work tomorrow, and have a look on a real osciloscope.
umejopa said:
It is such a stupid design, why connect the full battery voltage to vcc, when you could have just connected it to the boards 3.3V rail.
But this isn't what killed the programmer. I was just connecting the throttle out signal to the litle USB Programmer/SLOscope IO, and then things went wrong....
adrian_sm
1 MW
Solved. I had a bad ESC.
Put the modified watt-meter on my other bike with the known good 100Amp ESC, and was able to tune the throttle, arm the ESC, and have the watt meter control the throttle 8) .... But for some reason it doesn't read Amps while doing it. When I switch the throttle back to the old hall throttle & modded servo controller, then the amp readings are fine. :? Hope I didn't let my soldering iron stray somewhere it shouldn't have while removing the 4 resistors on PA4-7.
Anyway, really happy to have this running. Now I need to get it out on the road to try it out.
Oh and I noticed that the ESC puts out 5.5V not 5.0V, so this may have contributed to the demise of my little programmer/SLOScope thing.
And buy another ESC, but the 100Amp models are on back order. Maybe I should get a Castle Creations ESC so I can get data logging, and experience quality.
I don't need the HV models, as I will only be running 5-6s. I'll be controlling the throttle externally to control ramp rate, and max current. So was think maybe the Phoenix ICE 100.
Can anyone recommend the best place to buy them? The going rate on ebay is ~AU$115 delivered.
And do I need to buy the castle link USB thing separately to make use of the data logging?
Put the modified watt-meter on my other bike with the known good 100Amp ESC, and was able to tune the throttle, arm the ESC, and have the watt meter control the throttle 8) .... But for some reason it doesn't read Amps while doing it. When I switch the throttle back to the old hall throttle & modded servo controller, then the amp readings are fine. :? Hope I didn't let my soldering iron stray somewhere it shouldn't have while removing the 4 resistors on PA4-7.
Anyway, really happy to have this running. Now I need to get it out on the road to try it out.
Oh and I noticed that the ESC puts out 5.5V not 5.0V, so this may have contributed to the demise of my little programmer/SLOScope thing.
And buy another ESC, but the 100Amp models are on back order. Maybe I should get a Castle Creations ESC so I can get data logging, and experience quality.
I don't need the HV models, as I will only be running 5-6s. I'll be controlling the throttle externally to control ramp rate, and max current. So was think maybe the Phoenix ICE 100.
Can anyone recommend the best place to buy them? The going rate on ebay is ~AU$115 delivered.
And do I need to buy the castle link USB thing separately to make use of the data logging?
Good you have it working a bite.
Will the moded wattmeter read Amp now when you don't have it conectet for trottel control ?
It is not as simpel that you have curent flow wrong.
Hope you get current reading working so you can test the current regulating i hade made in the program. And change the the regulator parametrar .
I waiting soo to it will be warmer no it is -20GrC
/Jonas
Will the moded wattmeter read Amp now when you don't have it conectet for trottel control ?
It is not as simpel that you have curent flow wrong.
Hope you get current reading working so you can test the current regulating i hade made in the program. And change the the regulator parametrar .
I waiting soo to it will be warmer no it is -20GrC
/Jonas
Kepler
10 MW
adrian_sm said:
You can buy them here http://arkrc.com.au/products/010%2d...-Phoenix-ICE-100A-8S-(34V)-Brushless-ESC.html direct in Collingwood if you dont want to wait. Yes you will need the programming cable. http://arkrc.com.au/products/010%2d0079%2d00-Castle%2dLimk-quick-connect.htmlIts a must.
adrian_sm
1 MW
Yes.umejopa said:
No. Wish it was that simple.umejopa said:
It is difficult to tell without an oscilloscope and just a DMM, but I think it is noise from the throttle PWM signal affecting the shunt sensing voltages.
The voltages I read across the shunt are pretty low, and now that we have the PWM signal using the same ground, I think it is leaking through and disturbing the quite low shunt sensing voltages.
I tried very quickly and crudely wrapping the throttle out wires around a toroid magnet, and suddenly I did start seeing some readings, they still weren't quite right, but it was an improvement.
So if this is the issue, what is the best way to stop the noise on the current sensing. Caps, toroids, ....?
There is already a capacitor across the two current sense wires, labeled C1 on the wattsup schematic.
Or just implement toroids on both throttle in and out cables.
Noise suppression in electronics is not somehting I know much about.
[EDIT] I have also made the remote shunt mod, so my shunt is now on the end of a cable about 0.5m long. So this probably doesn't help.
adrian_sm
1 MW
Kepler said:
Thanks for the link.
How's your throttle interface going? Ready for some beta testers yet?
adrian_sm
1 MW
I had a few questions about the drive pop up in a PM, but saw no reasons to keep the answers private so I will post it here instead.
Most of the trick is in setting it up right once on the bike. You need to adjust the position of the drive on the seat tube, as this directly relates to the most important thing which is the depth the motor digs in to the tyre. This and the pressure in the tyre dictates the amount of contact force, which then dictates how much torque can be transmitted. The key other element is the coefficient of friction between the motor and the tyre. By using the grip tape/belt sander material we have increased the coefficient of friction, allowing a lower contact force with the tyre for the same torque. The benefit of this is less stress on the parts, and better efficiency.
It is assembled coaxially with the pivot axle. One end of the spring engages in a small hole in the pivot block, the other to the collate at the end of the pivot axle. This allows you to turn the collate and adjust the amount preload torsion the spring applies to the axle. So you can easily adjust the force to the geometry of your bike, and the weight of the specific motor used.
This image shows the relevant parts the best.
Have a look at this post for all the parts layed out. It should give you a pretty good idea of what it what.
Anyway I hope that answers the PM's questions, and might have helped someone else get their heads around the design.
Cheers,
Adrian
I just got it from the scrap bin at work. But you should be able to find it through a local plastics supplier. If you are lucky they might just have an off-cut they will give you. The material is acetal.
Don't worry too much about the theory. If you make the swing arm with a similar geometry (ie. distance between the pivot and the motor) you'll be fine.
Most of the trick is in setting it up right once on the bike. You need to adjust the position of the drive on the seat tube, as this directly relates to the most important thing which is the depth the motor digs in to the tyre. This and the pressure in the tyre dictates the amount of contact force, which then dictates how much torque can be transmitted. The key other element is the coefficient of friction between the motor and the tyre. By using the grip tape/belt sander material we have increased the coefficient of friction, allowing a lower contact force with the tyre for the same torque. The benefit of this is less stress on the parts, and better efficiency.
I haven't abused this thing enough yet to give solid recommendations. But Kepler has had good success with the Turnigy K-Force 100 Amp, and mine has held up without an issue.
The deadstops limit the angle which the swing arm (that the motor mounts on) can rotate relative to the clamp block that attaches to the seat tube. This allows you to easily tune the distance the motor sits from the tyre when not in use, and the how far it rotates when under load. Which sets the maximum depth the motor pushes in to the tyre. It makes it pretty easy to fine tune the drive to suit a range of different bike geometries.
Yes. It is vital to counteract the weight of the motor, to ensure the motor can lift in to and engage the tyre when you apply the throttle.
It is assembled coaxially with the pivot axle. One end of the spring engages in a small hole in the pivot block, the other to the collate at the end of the pivot axle. This allows you to turn the collate and adjust the amount preload torsion the spring applies to the axle. So you can easily adjust the force to the geometry of your bike, and the weight of the specific motor used.
This image shows the relevant parts the best.
Have a look at this post for all the parts layed out. It should give you a pretty good idea of what it what.
Anyway I hope that answers the PM's questions, and might have helped someone else get their heads around the design.
Cheers,
Adrian
billyjones454
100 mW
- Joined
- Feb 16, 2011
- Messages
- 44
Thank you for answering my questions! I love the simplicity and small setup size! Brilliant!
One more question - What is a good C rating for this motor, if I'm using 2x 5000mah 5s Turnigy LiPo's?
One more question - What is a good C rating for this motor, if I'm using 2x 5000mah 5s Turnigy LiPo's?
adrian_sm
1 MW
No worries.
By C rating for this motor, I assume you mean of the batteries. The short answer is the best you can afford.
But I am using 20C 5s5Ah in a 2p or 4p config.
At the moment there is no current limitting implemented, so I can see peaks of 2-2500W, which is 100-120 Amps.
For a 5Ah battery, that equates to 20C <= Bad
For a 10Ah battery, that equates to 10C <= Okay
For a 20Ah battery, that equates to 5C <= Good
So a 20c batteries in a 2p config should be fine for the batteries health. The thing I haven't got my head around yet is the ripple voltage effect on the ESC. As I have no way to measure it at the moment I have no idea how much of an issue it is. But I did kill my small cheap 85A controller, possibly while doing no load bench tests with a single 5s5Ah battery. So the battery C rating may be of more concern for the ESC's health.
Anyone with more experience in this area feel free to contribute.
By C rating for this motor, I assume you mean of the batteries. The short answer is the best you can afford.
But I am using 20C 5s5Ah in a 2p or 4p config.
At the moment there is no current limitting implemented, so I can see peaks of 2-2500W, which is 100-120 Amps.
For a 5Ah battery, that equates to 20C <= Bad
For a 10Ah battery, that equates to 10C <= Okay
For a 20Ah battery, that equates to 5C <= Good
So a 20c batteries in a 2p config should be fine for the batteries health. The thing I haven't got my head around yet is the ripple voltage effect on the ESC. As I have no way to measure it at the moment I have no idea how much of an issue it is. But I did kill my small cheap 85A controller, possibly while doing no load bench tests with a single 5s5Ah battery. So the battery C rating may be of more concern for the ESC's health.
Anyone with more experience in this area feel free to contribute.
adrian_sm
1 MW
Another PM question that others might be interested in.
ABS is a good engineering plastic, but not as strong a acetal in terms of yeild and impact. So if you can find it try and stick with acetal. You will need to go to engineering plastics suppliers, not your hardware store to find it though.
An alternative design is to use aluminium, and then use a bearing material for the shaft interface. Like Nyliners, bronze bushing, or real bearings if you get fancy.
ABS is a good engineering plastic, but not as strong a acetal in terms of yeild and impact. So if you can find it try and stick with acetal. You will need to go to engineering plastics suppliers, not your hardware store to find it though.
An alternative design is to use aluminium, and then use a bearing material for the shaft interface. Like Nyliners, bronze bushing, or real bearings if you get fancy.
adrian_sm
1 MW
Awesome. Thanks for that.
I did stumble across some link on the castle site that said something similar.
http://www.castlecreations.com/products/clink.html
That makes the buy decision easier.
I did stumble across some link on the castle site that said something similar.
http://www.castlecreations.com/products/clink.html
That makes the buy decision easier.
aaronski
1 kW
Hi Adrian,
I'm trying to replicate this design in CAD and machine one up, but, well, I'm not any kind of good at that. Do you have any plans to have these built and sell them?
I'm trying to replicate this design in CAD and machine one up, but, well, I'm not any kind of good at that. Do you have any plans to have these built and sell them?
