Much More Powerful Pie Now!!!

[Leslie]

Yeah take a big picture like you did and check the whole board via the PC.  It will take a while longer than it would just by eye.

This is the beauty of digital photography doing electronics.  Like I just did then, You spot see the smallest of problems.

[Leslie]

I designed a relay and auto precharge circuit that can take 24v to 60.

http://www.ficrelay.com.hk/details/FRA4.pdf

Can handle 7200 watts at 240v 30 amps AC,

The precharge is cute,

Battery and controller goes directly to the contact tabs of the relay.  A wire leads from the battery contact of the relay to an Lm338 to bump the voltage down to 48v to my relay coil delay circuit and to my lm2576 switcher regulator for my lights.

I used one poles throw of a DPDT rocker switch to bypass the relay contacts to send through a, resistor, thermistor setup 300ma to the controller and the other pole throw goes through a thermistor to a LM317t @12v to charge 2200uf cap around the relay coil at 12v  You must limit the volts to under 48v to the LM317t as it can only do a 37v drop over any voltage.


The cap charges slower so the rocker switch resistors to the controller has a chance to precharge the controller caps.  The thermistor to the LM317t limits the current surge that may result by charging capactor on the relay, must not allow more tahn 1.5 amps to pass the lm317t to the coil delay cap.  

I used a diode in somewhere to block something somewhere, I forget but if youre interested I can put some more effort into this tell you more about how it works.  

[o00scorpion00o]

I designed a relay and auto precharge circuit that can take 24v to 60.

http://www.ficrelay.com.hk/details/FRA4.pdf

Can handle 7200 watts at 240v 30 amps AC,

The precharge is cute,

Battery and controller goes directly to the contact tabs of the relay.  A wire leads from the battery contact of the relay to an Lm338 to bump the voltage down to 48v to my relay coil delay circuit and to my lm2576 switcher regulator for my lights.

I used one poles throw of a DPDT rocker switch to bypass the relay contacts to send through a, resistor, thermistor setup 300ma to the controller and the other pole throw goes through a thermistor to a LM317t @12v to charge 2200uf cap around the relay coil at 12v  You must limit the volts to under 48v to the LM317t as it can only do a 37v drop over any voltage.


The cap charges slower so the rocker switch resistors to the controller coil has a chance to precharge the controller caps.  The thermistor to the LM317t limits the current surge that may result by charging capactor on the relay, must not allow more tahn 1.5 amps to pass the lm317t to the coil delay cap.  

I used a diode in somewhere to block something somewhere, I forget but if youre interested I can put some more effort into this tell you more about how it works.  

Thanks 317537, I think for the moment I will just see how it goes with reliability. I resolderd the wire, was actually raised above the track altogether and would not have came in contact, You could only see it from an angle. Thanks for spotting the possibility  of a burned pie though!   

I'm dying to get on the bike again, but the weather is not so good, still windy and many heavy showers, and looks set to continue until next weekend, but hopefully I'll get out now and again.

[Leslie]

Sorry to put you through all that, it could be a lot worse if the solder was like the picture looked.

Time to get the wet weather gear out.

Clear safety glasses to stop rain pelting your eyes out.

water proof trousers.

Gum boots

Ski mitts are awesome in winter keep you fingers togethr and warm..

Rain proof jacket.

Any other ideas for some good wet weather apparel?

[o00scorpion00o]

Sorry to put you through all that, it could be a lot worse if the solder was like the picture looked.

Time to get the wet weather gear out.

Clear safety glasses to stop rain pelting your eyes out.

water proof trousers.

Gum boots

Ski mitts are awesome in winter keep you fingers togethr and warm..

Rain proof jacket.

Any other ideas for some good wet weather apparel?

Better to be safe than sorry, I really didn't notice it, I was surprised by how the camera takes good close ups.

I got a great cure for bad weather. called Indoors with a nice cup of tea!    The wind is a real pain here, but would be good for a wind turbine! We also get lots of rain, so I don't think I should ride much in the rain. I stick to my trusty 200000 mile audi diesel for winter lol

I am waiting for others to try the mod, Because I want to see what people think of the torque! I wonder how it compares to the likes of the bmc black lightning 1kw motor? or even the bmc v2 torque. But I'm well impressed with it! If the worst happens I will install an external controller. Maybe the magic controller.

where are you from 317537 if you don't mind me asking?

[Leslie]

Queensland Australia.

I never had a licence and I live 10kms ride to any shops with 7 other souls that rely upon me. so these bikes are the backbone to my life.

It doesnt get too cold down up here, cold enough to puts my great coat on and a couple of layers of under garments and a thick scarf.  Forget using pedals with all the gear on.  Maybe a little pedals to get the blood warm.

Any cold really gets amplified by the wind.

You just need to look at motor cyclist gear, it can be done.  

As far as I can tell your original shunt is in the wrong place.  I think at least it should go after the 63v cap because the fets are always draining the cap dry with every switching event and it is a small cap too.

all my other controllers negative - input feed go straight to the cap negative terminal then lead to a shunt then to the fets.  Not to the shunt, then to the cap, then to the fets.  

You can see in this picture the topography between your controller and a MC 36v earlier version.  These modifications mayber put inplace to add reliability to the controller under regen.  All this gets a little complicated and I need to work out how the changes alter the physics of the circuit.



Certainly that brown cap is very small and would possibly be effected by the switching load of the fet through the shunt into the larger cap and all depends on the smaller brown caps positive feed on the other side too.

On other magic controllers I have seen they use a much larger caps, where on your controller there is only a small cap in their place.

[Bikemad]

Certainly that brown cap is very small and would possibly be effected by the switching load of the fet through the shunt into the larger cap and all depends on the smaller brown caps positive feed on the other side too.

On other magic controllers I have seen they use a much larger caps, where on your controller there is only a small cap in their place.

Leslie, the Pie's controller has some of its capacitors mounted on the opposite side of the PCB:



The smaller brown one is probably fitted in addition to the large one on the rear.

Alan
 

[Leslie]

Certainly that brown cap is very small and would possibly be effected by the switching load of the fet through the shunt into the larger cap and all depends on the smaller brown caps positive feed on the other side too.

On other magic controllers I have seen they use a much larger caps, where on your controller there is only a small cap in their place.

The smaller brown one is probably fitted in addition to the large one on the rear.

Alan
 

Or probably not.  The 15 amp limit may indicate this is a magic controller that was made for the minimotor and the programing ability was there to allow the 24v operation and stuff from long ago..

Why the tiny little polyester cap in the first place if there is a huge electrolytic under it? There is no point unless youre talking low ESR tantalum voltage ripple filters. .

But alas you are right, unless we see the otherside of the board I can only speculate..

[Leslie]

I put some further thought into the other caps that may be on the other side,.

If there are other caps on the other side of that 63v electrolytic cap, the cap that on his board is glued straight over with no glue between solder joints of the suspected cap underneath.  Electrolytic cap cases are great for a short circuit when mounted on a PCB in such a way.  As I posted before the brown cap has no use with a larger cap ontop.  It would only add picofarads to the larger cap its in parallel with.

The evidence I would say is pointing towards that there is nothing underneath.

[Lightning Shadow]

Hey guys, In case you haven't seen the thread on ES about similar versions of these controllers.
It should handle 2000w ok, I tested up to 4800w without issue, eventually reduced the shunt resistance too much which killed it. If you do try it don't do it continuously and don't go higher, especially being inside the motor's case I'd limit it to around 3000w for normal road riding, but it depends on how and where you're riding, since it only peaks out at low rpm.

http://endless-sphere.com/forums/viewtopic.php?f=2&t=11015

I run my off-road bike with a controller that has 12 of the same fets that are inside magic controllers, and I push it to 5kw max from full to empty.

[o00scorpion00o]

hi again. Another update. I cooked my controller. I went up a fairly steep hill, It pulled like a train all the way up. Got to the top and stopped to check if I could feel heat. It was warm, good I said, Started off again then no power. 5 beeps. So I guess I cooked the controller  .

My god has this motor got some muscle. It is addictive, Correction, WAS!

I wonder how hard is it to order another controller? as they don't have any on the site. The external  magic controller pcb won't fit, because of the caps locations etc. The other thing I could do is run phase & hall wires through the axle and use the external magic controller. Does anyone now if I will get more top speed from the external?

[Leslie]

Doesnt 5 beeps mean controller went into LVC or something. Actually the documentation for that one reads unknown Email tom.

 Like 

Well ya warranty is void.  At least you got over that one pretty fast.


LOL

Remove the internal controller already.

Replace the fets there are only 6 of them.  Put in the 100v version fets and some 100v caps.

You can come up with a thermal solution now.

Maybe mount the controller fets onto a massive heat sink. It may suit you keep the fets cool.

[Leslie]

There are some better parts out there.

I found a replacement as it seems with better on-resistance and a higher voltage rate.  Give it a little time and we see if we can make the Pie magic again.

[o00scorpion00o]

Doesnt 5 beeps mean controller went into LVC or something. Actually the documentation for that one reads unknown Email tom.

 Like 

Well ya warranty is void.  At least you got over that one pretty fast.


LOL

Remove the internal controller already.

Replace the fets there are only 6 of them.  Put in the 100v version fets and some 100v caps.

You can come up with a thermal solution now.

Maybe mount the controller fets onto a massive heat sink. It may suit you keep the fets cool.

I remember replacing 2 fet's in the external controller. I remember they were a B***h to remove with all the solder on both sides.
If I remember correctly I thought those fet's were supposed to handle 140 amps?

Would changing caps to 100 volts matter much as my battery is only 52 volts?

I went up the hill at max throttle and the speed picked up quick, and it only pulled about 35 amps max, Max current I saw on take off was around 65 amps! As I said the pie was still working when I got off to check for heat, It's only when I went to pull off, I turned the throttle on and off a few times so as not to flip her over on the hill, ( seriously It had that much torque!) to try and take off easier, Maybe it didn't like that?

When the external controller died before, there was no power at all, and the battery bms tripped, This time there is power and 5 beeps, But the wheel is harder to turn, so I've no doubt there is fried fet's!

[Leslie]

http://www.datasheetcatalog.org/datasheet/stmicroelectronics/9231.pdf

Here are the fets I am pretty sure you have,

75v R-on 75mOhm 120 amps max



Or look at these ones.

http://www.st.com/stonline/products/literature/ds/8220/ste250ns10.pdf

100v R-on 55mOhm 240 amps max,

These fets allow you to run wires off you board and the wires can be screwed, your fet bridge made into a module.  Once you get your fets off the logic board you can go for all you like. Make it suck the earth's gravity into your pie and spit dirt out in return.

Replace all six fets with high voltage, high current wire. Now just hook those babies up to the right connections and by the power of luck it should work.