Can I Drive an H-Bridge with a "slave" H-Bridge

[DamoRC]

Know little about electronics but I am keen to learn more.

I am trying to come up with a solution for a radio controlled car whose stock electronics will handle a maximum of about 8V. I would like to use the stock electronics to drive an auxilliary H-Bridge that could handle a max of 15V (peak load of approx 20A). Stock electronics use an H-bridge to drive brushed motors.

There are a number of commercially available products that can do this.
One is from AtomicMods, the V2 (**broken link removed**). This reads the output from the leads that would normally go to the motor to drive an H-Bridge. Main advantage is that you don't have to modify the stock electronics to use it.

The other product is called Spider (**broken link removed**. This one requires you to remove the stock FETs and wire the H-Bridge gates on the stock electronics to the Spider (which is possibly a stacked FET H-Bridge).

Both products allow forward, reverse, braking and coasting.

I am wondering if there is a combination solution possible by simply taking the drains from the stock H-bridge and using these as the gates on an auxilliary H-bridge. Could the drain outputs (the motor leads) from one H-Bridge be used as the gate inputs for a second H-bridge? (I understand that there would have to be some logic involved to allow for the polarity changes as you go from froward to reverse).

Thoughts / Comments / Thanks.

 

[dknguyen]

You can't usually because a motor accepts 2 (or 3 wires) and an bare bones H-bridge accepts 4 (or 6) inputs. So the output of an H-bridge won't usually fit properly into the inputs of another H-bridge without some (rather pointless) circuitry that will basically read the signals from the stock drivers (without the power aspect that the stock driver is designed for in the first place) and convert it into something to drive another H-bridge. In my mind this is all pointless because you can just use the receiver to drive the auxillary H-bridge directly (ie. add circuitry on your aux driver to interpret the receiver signals). WHy bother making your receiver drive a stock H-bridge drive an auxillary H-bridge?

Just swap out the darn thing.

 

[mneary]

You need to detect more than polarity; you need to detect the non-driving state. By the time you've completed this, you have the circuit of your first option.

 

[DamoRC]

Okay:

You can't usually because a motor accepts 2 (or 3 wires) and an bare bones H-bridge accepts 4 (or 6) inputs.

So I may have understated my electronics knowledge in that I do understand the difference between the number of inputs to a motor and an H Bridge.

In my mind this is all pointless because you can just use the receiver to drive the auxillary H-bridge directly (ie. add circuitry on your aux driver to interpret the receiver signals). WHy bother making your receiver drive a stock H-bridge drive an auxillary H-bridge?

Unfortunately, your point of view on the "pointlessness" of this excercise is different from mine. Maybe I mis-stated the goal. The goal is to be able to deliver more V and A to the motor without modifying the stock electronics (which I failed to mention are not hobby grade and are, therefore, not easily "modified). This is obviously not entirely unreasonable as AtomicMods seem to be selling enough of these thing to warrant developing them and bringing out new versions. The advantage to this approach has little to do with electronics and more to do with flexibility. For example if you had multiple vehicles you could easily move the auxialliary board from one to the other returning the first vehicle to stock operation without too much work.

some (rather pointless) circuitry that will basically read the signals from the stock drivers (without the power aspect that the stock driver is designed for in the first place) and convert it into something to drive another H-bridge.

- is exactly what I am looking for.

Thanks

 

[dknguyen]

Okay:


So I may have understated my electronics knowledge in that I do understand the difference between the number of inputs to a motor and an H Bridge.



Unfortunately, your point of view on the "pointlessness" of this excercise is different from mine. Maybe I mis-stated the goal. The goal is to be able to deliver more V and A to the motor without modifying the stock electronics (which I failed to mention are not hobby grade and are, therefore, not easily "modified). This is obviously not entirely unreasonable as AtomicMods seem to be selling enough of these thing to warrant developing them and bringing out new versions. The advantage to this approach has little to do with electronics and more to do with flexibility. For example if you had multiple vehicles you could easily move the auxialliary board from one to the other returning the first vehicle to stock operation without too much work.

- is exactly what I am looking for.

Thanks

My point is the work you have to do to get something like that is more work than if you just built another one from scratch. Either way, you are building something that takes a signal, and drives a motor. NOw you could restrict yourself to make something that must use the signal from another driver which is not optimal for your purposes and build circuitry to "work around it", or you could or you could eliminate the "middle man" and make things easier and simpler.

BUt if it's not hobby grade then things get a bit harder, though I still don't see why you can't just remove the stock bridge from the truck unless it is welded onto the radio receiver (which needs to stay behind in the truck). WHat you would be building would be exactly the same as a normal run ifof the mill uC H-bridge with one small difference- the inputs to the bridge would be protected and have conditioning circuitry in order to accept the stock bridge's output which is probably of the wrong polarity and too higha voltage for the PIC. And then you'd just have code to interpret the states.

If you already have a barebones auxillary H-bridge, then the extra thing you would need is a microcontroller to tie together the two H-bridges. It's output drives the auxillary H-bridge transistors, and it's inputs are connected to the output of the stock bridge through conditioning circuitry. With code to interpret input to output.

When I said a motor accepts 2 wires and an H-bridge accepts 4 inputs, I meant this:

A motor accepts 2 wires, therefore an H-bridge has 2 outputs. But an H-bridge accepts 3 inputs. THis means that if you used the output of an H-bridge as signals to another H-bridge, you definately lose the ability to fully control the slave bridge since the master bridge is unable to represent as many states as the slave bridge can take on. The worst case scenario would be that the control signals cannot represent enough states to command the H-bridge's bare minimum funfctions, regardless of how much functionality you cut out in order to simplify the control signals needed.

You need at least 3 states (dropping all flexible control of the individual transistors in the aux bridge) which are forward/coast(or brake)/reverse. An h-bridge's output can represent a minimum of two states + polarity and -polarity. But you need a third state which is a problem. Unless the stock bridge is able to brake the motor (connect both terminals together, which by virtue of the H-bridge means connecting them both to either +V or GND), you have a state but this state represented by a floating voltage or high impedance state. It is a "lack of voltage". This will occur if the stock bridge cannot brake and only let the motor coast. You end up with your third state being an "absence of voltage" which obviously can't be read by the PIC.

If it does brake, you need to figure out whether the stock bridge ties the motor terminals to +V or ground. If it ties it to ground, great, you have a 3 voltage levels the stock bridge can output as signals (+, -, and 0V). If it ties it to +V, then you still have these states, but 0V is now only relative between the stock bridge's outputs and not the PIC's ground so you need to rereference it somehow.

One ways to do this is use optoisolators (and about the only way to isolate a DC signal really). Coincidentally, because optoisolators use pull-up (or pull-down) resistors on their outputs, this means that they are able to produce a voltage output in the absence of a voltage input which lets you turn the floating output of a stock-bridge that is only able to coast and not brake into a voltage allowing the PIC to read it. If you also make creative use of optos, they will also let you rectify the negative polarity signals of the stock bridge so that the PIC can read them. I believe the circuit is to have two sets of opto LEDs+ current limiting resistors connected in anti-parallel to each other. THeir outputs are connected to two separate pins on the PIC with one separate pull-up resistor for each pin.

Optos are also good in that they can handle the higher voltages outputted by the stock bridge (you just resize the current limiting resistor). I don't see a simpler way you could rectify and re-reference DC signals without optos. Other ways would include floating transformers with lots of refresh circuitry to preserve the DC characteristics (either as discrete components or as an iCoupler IC) which would need isolated floating supplies, or op-amps with bipolar supplies.

 

[DamoRC]

Thanks dknguyen.
I have a lot to think about in what you have said.
I was hoping to avoid the PIC route specifically due to the expense of the hardware needed to program it. I will need to detect 4 states (forward, reverse, coast and brake).
I will also need to learn a lot more about basic electronics before taking this on.

Thanks again for your detailed response.