Large power H-bridge
McFluffin, Fri May 26 2006, 05:57AMI have a large robot that I constructed a while back from some large 50V motors that I got at a flea market for $1 each. I took some pieces of scrap metal, a laptop, some relays, a wifi card, a camera, some car batteries, and threw together a robot that was large enough to be ridden on. This ran off of one car battery that was "dead." I eventually got 4 new car batteries from some Toyota Priuses and want to upgrade the power system to use a solid state design as well as be able to handle the increased power level. My meter can only read to about .1 ohm, so I guess that the resistance of the motors is around that, but I really have no idea. At their 50V rating, this would be around 500A(I measured .1 when the were in parallel with each other), which is quite a lot. I know that the motors are large, but not really sure if this is accurate. I might not be able to get that high, but I'd like to try to get as high as possible and will probably have to run it at lower voltage.
I was thinking that one could take IGBTs and put balancing resistors on them to make them evenly take the power. I have some 12 gauge magnet wire that I could use around 5 or 10 feet or so to get around .01ohm of resistance. This should be low enough to not snag all of the power from the motors, especially since there will be several of these resistors in parallel. I have several 200A IGBT bricks that I could use, but I'd preferably like to use some smaller packages since I won't be sad if I blow them up etc. I would start with a low power level and try to work my way up.
One large problem with the current design is that there is no gearing system on the motors. The wheels are directly attached, which can put a lot of strain on the motors. The robot is made out of wood(thats what was laying around), so the only weight is really from the batteries. I have been thinking about trying to use some bike sprockets, but haven't come upon any yet.
Any ideas/tips would be great.
Re: Large power H-bridge
Wilson, Fri May 26 2006, 11:20AM
I don't think there will be a constant 500A current draw, you would need ridiculously massive windings to handle that. Although, do expect a possible 500A current surge on start up. DC motors develop what is know as back emf, opposing the direction of electric current through them(related to lenz law). This back emf increases as the motor gathers speed, upon which the current draw will decrease accordingly.
So your H-bridge may need to handle maybe 400A for a second or 2, before the current drops back to something more reasonable.
A senior member should clarify this, i'm not too sure if i explained it too well, or even correctly to tell the truth
Wilson, Fri May 26 2006, 11:20AM
I don't think there will be a constant 500A current draw, you would need ridiculously massive windings to handle that. Although, do expect a possible 500A current surge on start up. DC motors develop what is know as back emf, opposing the direction of electric current through them(related to lenz law). This back emf increases as the motor gathers speed, upon which the current draw will decrease accordingly.
So your H-bridge may need to handle maybe 400A for a second or 2, before the current drops back to something more reasonable.
A senior member should clarify this, i'm not too sure if i explained it too well, or even correctly to tell the truth
Re: Large power H-bridge
McFluffin, Fri May 26 2006, 02:35PM
I was wondering that. How can I safely measure the current draw of the motors under load? My meter is only suppose to measure up to 10A, and I'm pretty sure its above that. I know that you can do it with an op-amp, but I don't really have a nice way to measure the response curve on that as I think I'd need a digital 'scope for that.
McFluffin, Fri May 26 2006, 02:35PM
I was wondering that. How can I safely measure the current draw of the motors under load? My meter is only suppose to measure up to 10A, and I'm pretty sure its above that. I know that you can do it with an op-amp, but I don't really have a nice way to measure the response curve on that as I think I'd need a digital 'scope for that.
Re: Large power H-bridge
..., Mon May 29 2006, 03:19AM
The best way to measure the current would be to use a 'current shunt'
It is basically a resistor, with a known amount (like 0.1ohm) of resistance which you measure with a volt meter, and use ohm's lay to calculate the current with.
To make one you just take a piece of wire and figure out the resistance of it (it is a good idea to make it a multiple of ten ohms, like .01ohm .001ohm, whatever). Then insert it into the circuit, and put your meter across it. Then you take current=voltage/resistance to calculate the current flowing.
..., Mon May 29 2006, 03:19AM
The best way to measure the current would be to use a 'current shunt'
It is basically a resistor, with a known amount (like 0.1ohm) of resistance which you measure with a volt meter, and use ohm's lay to calculate the current with.
To make one you just take a piece of wire and figure out the resistance of it (it is a good idea to make it a multiple of ten ohms, like .01ohm .001ohm, whatever). Then insert it into the circuit, and put your meter across it. Then you take current=voltage/resistance to calculate the current flowing.
Re: Large power H-bridge
HV Enthusiast, Mon May 29 2006, 12:40PM
Best way is to use a current transformer. You can build one (very easy) or buy small commercial units (Magnetek has nice small plastic ones for $6.00) which can measure peak current up to 500A.
HV Enthusiast, Mon May 29 2006, 12:40PM
Best way is to use a current transformer. You can build one (very easy) or buy small commercial units (Magnetek has nice small plastic ones for $6.00) which can measure peak current up to 500A.
Re: Large power H-bridge
Steve Conner, Mon May 29 2006, 01:31PM
It's DC current. So if you wanted to use a current transformer it'd need to be a Hall effect type.
I think gearboxes would be a pretty good idea :-o
Steve Conner, Mon May 29 2006, 01:31PM
It's DC current. So if you wanted to use a current transformer it'd need to be a Hall effect type.
I think gearboxes would be a pretty good idea :-o
Re: Large power H-bridge
McFluffin, Mon May 29 2006, 06:03PM
Odd that you say that...how about this for a current transformer:
I was never clear on this one...what is the 600V for and what is the 10kV for? I have a couple of smaller ones from a Prius and one made by someone else. I'll try using them to measure power output. I think the "other" one that I have was made for measuring DC.
Any suggestions on the H-bridge? I forgot to mention this, but is there any use in trying contactors?
Update:
Started building an H-Bridge using some IGBTs I had. Heres a pic of it starting out:
I tested it at 50V and ran the robot a couple of feet. No sign of any heating, so it should do pretty good. Maybe a little excessive using 4 IGBTs. They are Fairchild FGA50N60.
McFluffin, Mon May 29 2006, 06:03PM
Odd that you say that...how about this for a current transformer:
I was never clear on this one...what is the 600V for and what is the 10kV for? I have a couple of smaller ones from a Prius and one made by someone else. I'll try using them to measure power output. I think the "other" one that I have was made for measuring DC.
Any suggestions on the H-bridge? I forgot to mention this, but is there any use in trying contactors?
Update:
Started building an H-Bridge using some IGBTs I had. Heres a pic of it starting out:
I tested it at 50V and ran the robot a couple of feet. No sign of any heating, so it should do pretty good. Maybe a little excessive using 4 IGBTs. They are Fairchild FGA50N60.
Print this page