Modifying a shunt with controlled results

[John in CR]

If mentioned how to do this a couple of times, but took pics this time. I thought it deserved its own thread, since I've noticed a number of members burning up motors and controllers recently after modifying their shunts in an uncontrolled manner.

Simply adding solder to a shunt is a losing proposition. I've done it myself and created failures as a direct result. Solder has both a low melting point and a relatively high resistance, especially compared to copper. If you have a controller not easily programmed, you can still modify its current limits by modifying the resistance of the shunt. In the example below I was trying to slightly more than double my current by reducing shunt resistance by just over half, so I wrapped half of the shunt length in copper and soldered the copper to the shunt legs. Note that my shunt had 4 legs, so to make sure I had good contact between the copper and all of the shunt legs I wound the copper in an alternating manner so each wrap touched all legs.

I've done this several times and very thin copper is definitely the way to go, since it bends much easier and a controlled result ensured.

Before solder


After solder


Note that I probably took it a bit past half the resistance, since I wrapped copper a bit more than half of the shunt length.

Even if you have a programmable controller, you may want to do such a mod, which is my case. The infineon controllers have a fixed regen current of 10A. With high speed wind motors 10A is little and results in very light braking even at the high setting. I wanted to double the regen force and that's requires double the regen current, so I needed to fool the controller by modding the shunt resistance.

 

[liveforphysics]

Nicely done John.

 

[John in CR]

Nicely done John.

Thanks Luke,
Since those SevCons are so hard to program that I can't buy one from you, I gotta get to a similar point another way. Hopefully dual blowers venting a souped up 24fet including dual 10ga on all the wires is enough to do the trick of a reliable 150-180A battery side, 350-450A phase. That's why the shunt in the pic was modded, to get programmability that high. I gotta get Hubmonster on Super V as reliable as Blue, because the few rides I got in before controller failure were too addictive. Every ebike I have officially sucks now, so I'm Jonesin big time.

My stepping stone approach, from good to better bikes, is great in theory, but controllers as the weak link which are just "black boxes" to me, sucks big time.

 

[Hugues]

Hi John,

Interesting.

Could you roll back a little for me and explain how/why would someone need to do this ? I'm interested to learn how one can change the regen torque.

If you have a thread explaining more in details what is your mod affecting on the controller and motor

thanks

Hugues in CH

 

[mistercrash]

I know the question is more than two months old but I came upon this thread in a search for shunt mod and I tried it on my generic 15 fet controller on my Motorino scooter. I don't really know how it does it but the result is this. I go up a hill on a daily basis going to work, going up that long hill, the scooter would slow down and sustain a speed of just 25 kph all the way to the top. After I did exactly what's in the pics above to my shunts, I now go up that hill with a sustained speed of 37 kph. That is quite a difference for something that just cost a few minutes of my time. Torque is noticeably better off the line also so it's safer going through a 4 lane street at an intersection. So thank you to the OP for sharing this.

 

[John in CR]

Hi John,
Interesting.
Could you roll back a little for me and explain how/why would someone need to do this ? I'm interested to learn how one can change the regen torque.
If you have a thread explaining more in details what is your mod affecting on the controller and motor
thanks
Hugues in CH

Hugues,

Sorry I missed the question earlier. Our controllers estimate current for current limiting based on it's known shunt resisistance. When the shunt resistance is changed, the controller still thinks the old shunt is there, so lower resistance means more current flows than the controller thinks. eg Cut resistance in half and you get double the current without changing anything else. Before programmable controllers became available, modifying the shunt resistance was our only means to change current limits. This was typically done by adding solder, but it gave pretty unpredictable results. I tried the method above and it turned out to be pretty accurate.

Regen currents are limited by the controller, but even the programmable ones I've had, just have 2 settings, high/low, but with my high speed wind motors the regen current limits didn't apply enough braking force for my taste even on high. I reduced shunt resistance to fool the controller into allowing higher regen current. This was a programmable controller, and I already had high enough running current for the performance I wanted, so after reducing shunt resistance to get more regen power I had to turn the programmed current limits down by the same % that I reduced shunt resistance so my running current limits would be the same.

John

 

[John in CR]

I know the question is more than two months old but I came upon this thread in a search for shunt mod and I tried it on my generic 15 fet controller on my Motorino scooter. I don't really know how it does it but the result is this. I go up a hill on a daily basis going to work, going up that long hill, the scooter would slow down and sustain a speed of just 25 kph all the way to the top. After I did exactly what's in the pics above to my shunts, I now go up that hill with a sustained speed of 37 kph. That is quite a difference for something that just cost a few minutes of my time. Torque is noticeably better off the line also so it's safer going through a 4 lane street at an intersection. So thank you to the OP for sharing this.

Mistercrash,
Glad it worked for you. With that kind of performance increase be sure to monitor controller and motor heat, especially at the peak of your biggest hill. I've blown a number of controllers taking the mod too far.
John

 

[Spacey]

I'd like to go the opposite way, I have one of the Mac hi torque motors with a 9 Fet controller (think it's 9 Fet). It pumps out way too much power for the long range I need, I can't program it so I was thinking of cutting one of the 2 shunts.

Would this half the power which would be about what I need to make this the long distance 16Ah 15.5mph runner I need?

 

[John in CR]

I'd like to go the opposite way, I have one of the Mac hi torque motors with a 9 Fet controller (think it's 9 Fet). It pumps out way too much power for the long range I need, I can't program it so I was thinking of cutting one of the 2 shunts.

Would this half the power which would be about what I need to make this the long distance 16Ah 15.5mph runner I need?

Yes, cutting one of 2 shunt wires will cut the current limits in half, and if that ends up too low, you can use the method above to increase it back by the amount you want.

 

[Spacey]

Cheers for that

 

[Spacey]

Bugger...just opened it up and remembered why I didn't cut one of the shunts before....it only has one shunt. I suppose I could cut a part of the shunt out and re connect it.

 

[John in CR]

Bugger...just opened it up and remembered why I didn't cut one of the shunts before....it only has one shunt. I suppose I could cut a part of the shunt out and re connect it.

Actually, you would need to lengthen it not shorten it. What you want is greater resistance over that span. If you have a blown controller, recycle a shunt from it. Then desolder one of the ends of the existing shunt and bend it up and out of the way. Solder in one end of the scavenged shunt, and then connect the 2 open ends to create a much longer shunt.

I've heard of others using other types of metal wire, but the metal for shunts is special. Copper is much lower resistance, but the resistance of the metal used in shunts changes little with temperature.

 

[Spacey]

Top advice John I have a few blown controllers kicking about somewhere lol.

 

[Spacey]

Now I'm confused...to reduce the power surely shouldn't I be shortening the single shunt? Just trying to understand the process.

 

[bigmoose]

Ok, I see you guys soldering shunts and trimming things. Attached is a dither of some controller schematic that I snagged off the board, and how the bigMoose would trim current sense. I would not be soldering shunts. I would just add a small 10 turn trimpot, perhaps 10 to 50 ohms across the shunt and then I would have an easily adjustable manipulation of the current sense.

Just cut the trace that goes to the comparator, op amp or the micro controller. Solder in the trimpot across the shunt and dither away. If you set the wiper half way you get double the current. The rest is proportional. Just be careful to not go all the way against the wrong stop or you get "infinite" current limit!

 

[John in CR]

Thanks Bigmoose,

If I understand correctly I could use this method to get a multi-way switch or pot outside of the controller to change my current limits at will. Do you see any risks doing this other than a short causing too little resistance , eg what if the connection is broken?

Instead, what about cutting the sense trace and putting the trimpot (later a multi-way switch with different resistances) in series there? Then we avoid going higher than programmed resistance, since the potential of going too high and immediately blowing the controller is a big concern for me.

Thanks so much for laying this out. I've only been asking for exactly this from those selling controllers for over 2 years.

John

 

[bigmoose]

John the failure modes are likely to fail to infinite current for most scenarios. Use a good quality 10 turn trim pot and solder it in the case and I think you will be fine. Fine wires to a handlebar mounted carbon pot may not be the best idea.

Think of the trimpot as two resistors doing the chore of a voltage divider across the manganin shunt resistor. You are setting the division ratio by where the pot is set. You could do this with two fixed resistors if you wished. I used this implementation when I designed a CNC stepper motor driver 20 years ago and could not find the right size shunt resistors for the current I wanted. In my case I calculated the required division ratio and used two fixed resistors.

You want the trimpot to be a low resistance to keep the source impedance low, because some controllers may do voltage division and filtering later. If you use 1K you may start changing their set points and filter frequencies.

 

[John in CR]

I think I get how it works now, so if I used 2 fixed resistors of equal resistance, one to each side of the shunt in place of the trimpot, then I'd get double whatever the current settings are pre-mod. Lower resistance on the battery side increases it to more than double, and higher resistance decreases it. If I put a switch to be able to cut the resistor circuit that connects to the battery side of the shunt, then I get pre-mod current limits with that part of the circuit interrupted.

I like the fixed resistor route, because it achieves a known result, and you're right I simply don't run battery voltage up to the handlebars. I'd probably just go with the switch on the controller. I've just wanted the ability to change my current limits without attaching to a computer to reprogram the controller. Assuming I can actually get in and find the correct trace and location to tie into the board, then this looks like a good solution to easily give me 2 current limits while away from home, one for normal riding and hill climbing, and a higher one for showing off on flatter terrain.

Thanks again,

John

 

[bigmoose]

John, I would explain it in words, but I would foul it up! Attached is another drawing with the fixed resistors labeled and an equation. Just keep the parenthesis straight and it will calculate out OK. Good luck!

 

[sn0wchyld]

Ok, I see you guys soldering shunts and trimming things. Attached is a dither of some controller schematic that I snagged off the board, and how the bigMoose would trim current sense. I would not be soldering shunts. I would just add a small 10 turn trimpot, perhaps 10 to 50 ohms across the shunt and then I would have an easily adjustable manipulation of the current sense.

Just cut the trace that goes to the comparator, op amp or the micro controller. Solder in the trimpot across the shunt and dither away. If you set the wiper half way you get double the current. The rest is proportional. Just be careful to not go all the way against the wrong stop or you get "infinite" current limit!

maybe i'm missing something here, but doesnt the full current to the motor go through the shunt? and if so, wouldnt adding a resistor in parallel mean that resistor will see (at times) the full battery voltage? so a 10ohm resistor might see 100v, thats 1kw of heat right there. i figure i am missing something if you have done similar before bm, but I just don't see where...

also, if you're worried about hitting infinate current, could you just add a low value resistor in series with the trim pot that'd prevent you hitting '0' ohms?

Otherwise, sounds like a great way to limit your power on the run... it'd be easy to have a small knob on the outside of the controller for easy adjustment.

 

[John in CR]

There's high resistance downstream, so high current will never go through this circuit.

Yes on the extra resistor as protection. Once I get around to trying this, I plan to use only resistors and have 3 switchable settings. I'd love to get it up on the handlebars, but I don't know about running pack voltage up there through some small wires.

John

 

[sn0wchyld]

There's high resistance downstream, so high current will never go through this circuit.

Yes on the extra resistor as protection. Once I get around to trying this, I plan to use only resistors and have 3 switchable settings. I'd love to get it up on the handlebars, but I don't know about running pack voltage up there through some small wires.

John

i thought the purpose of a shunt was to measure the voltage drop across a known resistance in order to determine the current? so battery amps has to flow though the shunt for it to be useful? or are my wires crossed...

another thought on this RV mod, is there any harm in chaining your effective shunt R it on the fly? if not, how bout using this as a simple variable regen brake?

you could use a break cable to control a RV pot form the handlebars, you'd need a small switch on the leaver to 'engage' the brake, and at the same time to switch a relay or micro-switch so your controller will 'see' a second RV pot with the resistance maxed out (rather than whatever setting you have for normal acceleration). with high reistance the controller would initially think it's hitting 10A or whatever regen when its really only 1A, but as you pull on the leaver it turns the RV pot, lowers the effective resistance of the shunt, and you're controller will then think its hitting 10A regen when in fact its a sliding scale up to 15 or 20A (or whatever you set it up to max out at). there's the obvious danger that an incorrect setup will be to hard on or even kill batteries and / or controller... but seems to me to be a fairly simple mod to enable variable regen on pretty much any controller with regen?

 

[Arlo1]

Dave Thats so Awesome! Man and so simple Why did I never think of that. I guess I might have if my mind wasn't on software for the last 2 years! Cheers Very good mod. And after seeing that I will never do it the other way again!

 

[Arlo1]

There's high resistance downstream, so high current will never go through this circuit.

Yes on the extra resistor as protection. Once I get around to trying this, I plan to use only resistors and have 3 switchable settings. I'd love to get it up on the handlebars, but I don't know about running pack voltage up there through some small wires.

John

i thought the purpose of a shunt was to measure the voltage drop across a known resistance in order to determine the current? so battery amps has to flow though the shunt for it to be useful? or are my wires crossed...

another thought on this RV mod, is there any harm in chaining your effective shunt R it on the fly? if not, how bout using this as a simple variable regen brake?

you could use a break cable to control a RV pot form the handlebars, you'd need a small switch on the leaver to 'engage' the brake, and at the same time to switch a relay or micro-switch so your controller will 'see' a second RV pot with the resistance maxed out (rather than whatever setting you have for normal acceleration). with high resistance the controller would initially think it's hitting 10A or whatever Regen when its really only 1A, but as you pull on the leaver it turns the RV pot, lowers the effective resistance of the shunt, and you're controller will then think its hitting 10A regen when in fact its a sliding scale up to 15 or 20A (or whatever you set it up to max out at). there's the obvious danger that an incorrect setup will be to hard on or even kill batteries and / or controller... but seems to me to be a fairly simple mod to enable variable regen on pretty much any controller with regen?

What DAVE just showed is a trim pot that essentially connects the shunt sense wire of the controller to a spot you pick and can vary from one end of the shunt to the other the power will still flow though the shunt but what the Brain of the controller sees can be varied! You could even just use a circuit to make an electric NOS on a stitch with a resistor or a pot if you want! Very cool. And if you crank up the power then set it down in the programing it will give you the same acceleration with stronger Regen! Or.... How about this... We set up a trim pot hooked to the brakes then as you squeeze the brakes it can use the circuit Dave showed us to make variable Regen!
Man that's awesome I almost want to play with a China controller again lol.

 

[sn0wchyld]

There's high resistance downstream, so high current will never go through this circuit.

Yes on the extra resistor as protection. Once I get around to trying this, I plan to use only resistors and have 3 switchable settings. I'd love to get it up on the handlebars, but I don't know about running pack voltage up there through some small wires.

John

i thought the purpose of a shunt was to measure the voltage drop across a known resistance in order to determine the current? so battery amps has to flow though the shunt for it to be useful? or are my wires crossed...

another thought on this RV mod, is there any harm in chaining your effective shunt R it on the fly? if not, how bout using this as a simple variable regen brake?

you could use a break cable to control a RV pot form the handlebars, you'd need a small switch on the leaver to 'engage' the brake, and at the same time to switch a relay or micro-switch so your controller will 'see' a second RV pot with the resistance maxed out (rather than whatever setting you have for normal acceleration). with high resistance the controller would initially think it's hitting 10A or whatever Regen when its really only 1A, but as you pull on the leaver it turns the RV pot, lowers the effective resistance of the shunt, and you're controller will then think its hitting 10A regen when in fact its a sliding scale up to 15 or 20A (or whatever you set it up to max out at). there's the obvious danger that an incorrect setup will be to hard on or even kill batteries and / or controller... but seems to me to be a fairly simple mod to enable variable regen on pretty much any controller with regen?

..... Or.... How about this... We set up a trim pot hooked to the brakes then as you squeeze the brakes it can use the circuit Dave showed us to make variable Regen!
Man that's awesome I almost want to play with a China controller again lol.

isn't that what I just said? makes me a bit sad that I'm getting rid of the hub, variable regen would be tits!



anyway, not sure what I was thinking with the shunt... the RV is only going to have the voltage of the shunt across it, not the entire pack voltage like I was picturing...