Interesting outcomes using voltage boosters - Turbo for bike

[cj7hawk]

Hi All,

Just thought I'd share something - I've been experimenting with voltage boosters with some interesting outcomes.

First the background. Due to some momentary stupidity, I managed to convince myself that if I got a 48v 500w brushless hub motor and ran it at 24v, I'd get around 250w output, because I never multiplied the voltage by the amps in my calculations - Both were correctly calculated - I just made the assumption that it would be half and forgot to check. Stupid me. The result is I got 125w, which while meeting the 200w legal limit in my country, was well below it.

The next issue was that although that was a good theoretical power, it did go slightly over that in practice, the back-emf limited my maximum speed to around 19kph. At 20, I could not put power into the motor at all, meaning I had a 19kph top speed. Given I should be able to do about 24, I thought on how to correct this.

The result? I bought a cheap $10 600w booster off Ebay and wired it up in series with the battery, set the output voltage to 36v and max power to 200w, then tried it out.

Great results - maximum speed now increased to 24kph before equilibrium and depending on where I set the maximum voltage, it continued to provide power above 24kph until back-EMF levels reached the voltage I had set, meaning I could go faster downhill or with a tailwind -

Also, it limited the drain on the batteries ( in terms of amps ) because I could overvoltage the motor, but the output was current controlled, so as a result, if I used too much current, the voltage dropped and it protected the input from overcurrent situations.

Now these things aren't perfect - they run at around 95-96% efficiency, but the end result is I now have a "Turbo" option on my bike and I'm working to fit a permanent solution with a turbo/bypass option to either go to battery direct or use the power from the booster and raise my top speed for a nearly equivalent cost in watt-hours used to achieve it.

It's probably not a big deal in the US, because you have pretty good bikes there, but here with a 200w limit, every bit counts. Meanwhile, it allows me to set a maximum power output from my motor, which would not be possible with a direct battery connection alone.

This information is just experimental - and I'm still collecting data on it, but thought I'd share with the forum. The extra losses for me in terms of range aren't a problem because I have a range-extended electric bike, so I can just fire up the onboard generator and run at full speed. Also, because of legal limits here, I can tweak the output to get the maximum power possible while maintaining a legal bike.

Anyway, they aren't a great solution for a lot of reasons, but this one did work and I was surprised at how well.

Some pics - ( yeah, kinda dodgy implementation I know ).

The booster itself - Typically cost around $10 to $15 on Ebay.

How I wired it up... I'll do better next time and make a nice housing for it - with a turbo/bypass switch.

Regards
David

 

[MadRhino]

Interesting.

Yet, if your bike wasn't looking like a time bomb on 2 wheels, maybe you could get away riding with much higher power.

 

[amberwolf]

A little coroplast from a political election sign after those are over would fix the look right up. (if you use one from the winner you may not have to paint it over )

 

[skeetab5780]

Have not seen a Huffy in a long time. I used to own one until my dad ran it over backing out of the driveway 20 years ago, which forced him to buy me a GT Palmer. This bike was much better but it made me realize how much i hated grip shifters. I still despise them to this day.

 

[cj7hawk]

Interesting.

Yet, if your bike wasn't looking like a time bomb on 2 wheels, maybe you could get away riding with much higher power.

Ow, kinda cruel - I'd criticize, but you're right - I'd better make sure the wiring is fixed up before I go riding, especially on ANZAC day... Wouldn't want to end up at the wrong end of a cop with bad intel.

I haven't printed the cover yet - I'll assembled that tonight and run the wires by a neater path.

Given I ride with a petrol motor attached to it most of the time, I'm going to get attention anyway. The police don't really have much idea about what to do when petrol engines are attached to electrical bicycles, and while legal, they do tend to think they've found an unlicensed motorcycle.

David.

 

[dnmun]

why not just use a 36V battery instead and skip using the DC/DC converter? that thing must get so hot it will cook you and it may be hot enuff to melt the coroplast too.

 

[cj7hawk]

why not just use a 36V battery instead and skip using the DC/DC converter? that thing must get so hot it will cook you and it may be hot enuff to melt the coroplast too.

Because
a) Total heat output from the device won't exceed 10w... And I can't feel any heat buildup at all.
b) 36v means using expensive batteries, which are not needed, and I would still need a down-converter to run at 200w @ 30v
c) All modes of failure of down converting would be illegal in Australia

Using a 36v battery is a solution to a different problem.

Regards
David

 

[Doctorbass]

In the past i have used the TYco and Arthesyn DC-Dc of 150W 30A 5VDC with 36-72V input to boost my ping battery of 48V to 60V. I just connected many of these 5V output DC-DC in serie with the battery output and connected all the DC-DC input to the battery 48V output. It worked like a cahrm for current under 35A =)

these dc-dc cost about 5-10$ each so each time you boost by 5V and 30A max it cost 10$ per 5V extra.

Doc

 

[riba2233]

why not just use a 36V battery instead and skip using the DC/DC converter? that thing must get so hot it will cook you and it may be hot enuff to melt the coroplast too.

Because
a) Total heat output from the device won't exceed 10w... And I can't feel any heat buildup at all.
b) 36v means using expensive batteries, which are not needed, and I would still need a down-converter to run at 200w @ 30v
c) All modes of failure of down converting would be illegal in Australia

Using a 36v battery is a solution to a different problem.

Regards
David

a) if you trust them.. but they lie
b) they are not expensive, voltage doesn't have anything to do with price, you wouldn't need "down-converter" because your motor controller is that, and it can also limit current, thus power
c) as I've said, your controller is buck converter.

 

[macribs]

In the past i have used the TYco and Arthesyn DC-Dc of 150W 30A 5VDC with 36-72V input to boost my ping battery of 48V to 60V. I just connected many of these 5V output DC-DC in serie with the battery output and connected all the DC-DC input to the battery 48V output. It worked like a cahrm for current under 35A =)

these dc-dc cost about 5-10$ each so each time you boost by 5V and 30A max it cost 10$ per 5V extra.

Doc

If there was a way to have a DC DC converter placed between the controller and the motor people could be free of the 20s limits (jailhouse) for most controllers. And if you could switch it on/off on the fly you would have an overdrive mode upping the voltage and RPM when reaching speeds close to max. Should give you more speed. And might even allow for an even lower gearing for better climbing and acceleration without any top speed sacrifices.

That would give controllers without FOC / OVS a different take on the extra speed but would as the other result in higher top speed even with lower cost controllers.

Is this possible?

 

[TheBeastie]

In the past i have used the TYco and Arthesyn DC-Dc of 150W 30A 5VDC with 36-72V input to boost my ping battery of 48V to 60V. I just connected many of these 5V output DC-DC in serie with the battery output and connected all the DC-DC input to the battery 48V output. It worked like a cahrm for current under 35A =)

these dc-dc cost about 5-10$ each so each time you boost by 5V and 30A max it cost 10$ per 5V extra.

Doc

If there was a way to have a DC DC converter placed between the controller and the motor people could be free of the 20s limits (jailhouse) for most controllers. And if you could switch it on/off on the fly you would have an overdrive mode upping the voltage and RPM when reaching speeds close to max. Should give you more speed. And might even allow for an even lower gearing for better climbing and acceleration without any top speed sacrifices.

That would give controllers without FOC / OVS a different take on the extra speed but would as the other result in higher top speed even with lower cost controllers.

Is this possible?

I guess anything is possible. I often day dream about things I want in my controller that dont exist, and I guess its all just market and cost and the fact its safer for them to just make the basic controller without having to worry about the %1 wanting some freak feature.

I wouldn't mind a controller that has USB out 5volts suppy or just a DC step down out for power to my lights and my mobius HD recording camera. Also wouldn't mind if the controller also had bluetooth so if the ebike cuts out randomly and I am not sure why (thinking its bad signal from my magnet break sensors) I could just load up the android/iOS controller app my on phone and check/disable break sensors to help diagnose my problem.

Features I want but might not be willing to pay for etc are just the tip of the iceberg for ebike controller manufactures I would guess.

I do like the idea of a step up DC turbo booster from a geeky point of view, I am thinking it might be at its most useful if your going down a hill and want to go faster, you don't need much power in terms of torque (current) you just need more force (voltage) to push down the hill faster.. I guess its a bit like the constant flow of water from a hose, if you can squeeze the nozzle and make the water turn into a jet stream you have increased the force (voltage) and then can aim it at a little rock to make it go down a hill just like making your ebike go down a hill..

 

[cycborg]

If there was a way to have a DC DC converter placed between the controller and the motor...

Is this possible?

The thing about DC-DC converters is that they convert DC to DC. The voltage between the controller and motor is pretty much the antithesis of DC, with PWM switching superimposed on commutation switching.

 

[macribs]

If there was a way to have a DC DC converter placed between the controller and the motor...

Is this possible?

The thing about DC-DC converters is that they convert DC to DC. The voltage between the controller and motor is pretty much the antithesis of DC, with PWM switching superimposed on commutation switching.

Too bad really. I was hoping for a en easy way around the 20s jail
So how does ie the adaptto's OVS mode give extra speed when engaged? Is it done just by increasing the current?
That will lead to hotter running motor as well as controller.

Is there any type of converter that can handle pulse width modulation so that converter can be placed between controller and motor?
It is the Fet's that are going on off creating the pwm? Well if such a thing does exist I guess it would be both complex and expensive as it would need bigger more expensive FET's then the original controller.

 

[riba2233]

Between motor, you would need a big transformer, and that's impossible of course. You just have to change your way of thinking, 20S isn't any kind of jail. You can have whatever power you want at 20S.
OVS is field weakening, it increases current in d direction.

It's actually pretty complicated, look:

http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=0CCcQFjAB&url=http%3A%2F%2Fprojekter.aau.dk%2Fprojekter%2Ffiles%2F17643253%2FPED4_1038C.pdf&ei=Skg6VbD3CMrnasfXgLAC&usg=AFQjCNHqf5h49chA91YMCvODmlNeh8q5Yg&sig2=0iFX1VCT1Z4CwAsPVTcQMw

Chapter 3.5

 

[cj7hawk]

why not just use a 36V battery instead and skip using the DC/DC converter? that thing must get so hot it will cook you and it may be hot enuff to melt the coroplast too.

Because
a) Total heat output from the device won't exceed 10w... And I can't feel any heat buildup at all.
b) 36v means using expensive batteries, which are not needed, and I would still need a down-converter to run at 200w @ 30v
c) All modes of failure of down converting would be illegal in Australia

Using a 36v battery is a solution to a different problem.

Regards
David

a) if you trust them.. but they lie
b) they are not expensive, voltage doesn't have anything to do with price, you wouldn't need "down-converter" because your motor controller is that, and it can also limit current, thus power
c) as I've said, your controller is buck converter.

Fair enough, and valid concerns, however I have to say I'm surprised by the testing, which has shown the booster to be quite a viable unit. Keep in mind I have 24v of 100A capable batteries - 2400w maximum output and I'm using up to 15A reliably and continuously. Lead Acid works well for my current application in that respect - but voltage here is limited by the size of the batteries. 7.2Ah is the compromise. In the case of Lithiums, perhaps you can tell me what that should be and what it would cost? - but whatever I use must be able to support charge currents at up to 15A for 26v output. ( charge, not discharge ). If the booster wasn't viable, I'd probably just have remained with 24v.

I'm not saying there aren't other options - but I wanted to try to address the problem and gave the booster a shot - and was relatively surprised it actually worked. For someone with batteries already, if more voltage is needed, why not use them? $10 is a pretty small price to pay, and I can drive the circuit to 40v with current limited to get even higher maximum speed to take advantage of downhill and tailwind situations -

Anyway, while variable buck converters do exist, but if one failed closed, then the output would break local regulations. So easier to start with a lower voltage and boost it.

Regards
David

p.s. Update pic now the circuit is housed.

 

[cj7hawk]

Between motor, you would need a big transformer, and that's impossible of course. You just have to change your way of thinking, 20S isn't any kind of jail. You can have whatever power you want at 20S.
OVS is field weakening, it increases current in d direction.

It's actually pretty complicated, look:

http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=0CCcQFjAB&url=http%3A%2F%2Fprojekter.aau.dk%2Fprojekter%2Ffiles%2F17643253%2FPED4_1038C.pdf&ei=Skg6VbD3CMrnasfXgLAC&usg=AFQjCNHqf5h49chA91YMCvODmlNeh8q5Yg&sig2=0iFX1VCT1Z4CwAsPVTcQMw

Chapter 3.5

There's no reason I suppose that you couldn't just use the motor controller to generate the signals and rebuild the drive section to take whatever voltage/current you wanted. Such would just need some decent FETs and heatsinking - And you could run off of any voltage you wanted to - either straight battery voltage or a higher application -specific voltage with boosters.

I mean, theoretically, you could probably run a motor at well over 100v, but after 48v, you do start to get into higher voltages so there are practical dangers.

Still, at some point you'd be using mains voltage inverters as the boosters in three phase - they would be able to provide thousands of watts.

Regards
David

 

[ecycler]

In the past i have used the TYco and Arthesyn DC-Dc of 150W 30A 5VDC with 36-72V input to boost my ping battery of 48V to 60V. I just connected many of these 5V output DC-DC in serie with the battery output and connected all the DC-DC input to the battery 48V output. It worked like a cahrm for current under 35A =)

these dc-dc cost about 5-10$ each so each time you boost by 5V and 30A max it cost 10$ per 5V extra.

That is really interesting. Did you document any of these experiments? I would love to learn more about how this went!

cj7hawk, why not try adding another and see what happens when you put them in series for 72v!!! hehe

 

[Doctorbass]

Yes i did.. but i just can't find the thread where i documented that.

the goal is to use a dc-dc that isisolated between input and output. So you take the battery and connect it to the input to power the dc-dc.. then you take the output of that same dc-dc and connect it in serie with the input (or battery witch is the same) then you get both voltage summed :wink:

In my exemple i took a 48V to 5V 30A dc-dc and connected the 5V output in serie to the 48V input to get 53V.. and you can scale that infinite time until you reach the max isolatio voltage between the input and output witch is often thousand+ volts.

It is not a magnet motor style prophetie... it dont create energy.. it just take few amp from the 48V and gove 30A at 5V output that is then in serie with the 48V.

I also bought multiple 250W 48V dc-dc. I have about 10 of these. My goal was to make a powerfull 96V battery from a 48V one by using multiple 48V to 48V dc-dc in parallel to build up the current ( 10 x 250W = 2500W of additional 48V so the total allowable power woul have been 5kW at 96V from a 48V battery.

But as well it is sometime simpler to just use the right motor winding and wheel size to get the desired speed... but for those who want to modify existing setup this is a great idea!

here is an exemple of my 58V to 68V booster from 150W 5V 30A dc-dc: ... and the pdf of the other dc-dc i have of 48V to 48V isolated and 250W power.

(the 48V dc-dc normally take between 36 to 75Vdc at the input.)

The most common dc-dc brand are Lucent, Arthesyn, tyco, or Vicor.

 

[cj7hawk]

Ahh, telecommunications use - That makes sense.

I've ordered a new booster to try that has digital controller so I can set the output exactly for voltage and current. These cost around $30ea so a little more expensive. But they claim 400w and have two coils, so I think they're balanced at about 200w each on the same circuit. Max input 15A and max output 10A. Should be able to handle the 300W maximum I'm planning on needing from my 24v system.

I've found the DC capabilities of the system are pretty useful. If I sit a little slow, I get good economy and boost gives speed - a perfect combination for a system that uses petrol to enhance range, so I can maximum the small batteries for distance when needed.

Completely isolating the lines is challenging - You'd need to decouple it with a transformer. Perhaps a very high frequency transformer could be smaller with less windings. Decoupling the output would be useful in terms of being able to build a better charger.

Though I'm paying $60 for a full set of replacement batteries, and they should last years with the generator boosting them. So while it's tempting to start working on that solution, I haven't been able to find the price justification.

Thanks for the details - An interesting approach.

David.