Anyone know of an 8channel H-bridge?

Go To Last Post
22 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Wahoo! First post!

To be ontopic: Does anyone know of any 8 or more channel H-bridge, preferably serially controlled, with each output giving 1A or more? Preferably NON SMD, but I think I'll manage even simpler SMDs.

For a robot. :-)

David

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

the heatsink should be HUGE! So probably no ;)
But I don't know everything...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You can get a dual motor controller that's serial based, but nothing beyond 2 motors. You could use a bunch of duals on the same serial channel with something like Pololu's controllers. What are you making that needs 8 motors?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Burn out just one of those 8 and then what do you do?Replace the whole thing? I bet they wouldn't be cheap either if they existed.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Think modular. Small/medium/large controller design.

8 amps into one device, 8 motors, 8 control ports, very hard to keep the pcb size down.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Chip or PCB/module?

Do you need to have independent control of all 8 motors? You can drive 2 motors with only 3 half bridges, but they can't run independently completely.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Not 8 motors! Just 4! 8 channel as in 8 half bridges.

Quote:
Think modular.

Think small. ;-)

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Two of the classic L293B? Add something like a tiny2313 to control it and you have a nice small serially
controlled 4 DC motor driver module.

What's the maximum size?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

How much more then 1A?

A few VNS1NV04D + ATTiny

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Again, the LMD18201 full H-Bridge is a rock solid device. I've used it in a couple of designs. There good up to 50VDC @ 3 amperes and, they are heat-sinkable.

Unless you use one of the Allegro devices, expect to use a heatsink. And, even then, the Allegro devices have wings protruding out from pins 3, 4 & 5 on one side and, pins 10, 11 & 12 on the other side, that get soldered to a large copper on the PCB area acting as a heatsink.

Either way you go, you'll more then likely need a heatsink.

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

8 channel pwm controller.... (close)
http://www.awce.com/pak5.htm

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

THanks all for the replies. I think I'll use a few small FET tranzistors. And a bit of logic. With good board design it could fit onto a small area.

Thanks.

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There are dual FETs in small paks... dual N or P or both... saw them in the digikey catalog

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

daqq wrote:
THanks all for the replies. I think I'll use a few small FET tranzistors. And a bit of logic. With good board design it could fit onto a small area.

Thanks.

The thing to be mindful of when using descrete FETs in an H-Bridge is, controlling some amount of dead time when cross switching the FETs.

If you turn on say, the top FET on one side of the bridge before the bottom FET on that same side fully turns off, there will be a direct short across the power source. There needs to be a certian numner of nS/uS delay after turning off one FET and turning on the other FET in the respective leg of the H-Bridge.

Also, a descrete FET H-Bridge will take up more realestate then an integrated device of an equivelent current rating, with far less functionality. Clear examples of this are the Allegro and TI H-Bridge packages (and others). Their integrated H-Bridge devices will offer benifits, such as current limiting, micro stepping control, thermal status, etc...

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Whats the next step up from h-bridge ICs that seem to poop out at about 50V and an amp or two? What if you wanted to build an electric scooter or something that used big V and A? Make your own controller from IGBTs? Whats the biggest amps on a TO220 type pack youve seen (big heat sink assumed here)

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have several IBGTs in a fairly large package. I don't know the proper designation for the package though. I got them out of some older Variable Frequency Drives (VFDs). Their rated at 20 plus amperes @ 480VAC.

I was toying with experimenting with them, using my 13.8V, 50 ampere power supply. I'm sure I could develop some nice torque then.

I have used some stepper motor drives from Gecko Drive. They are rated at 7 amperes between 24VDC & 80VDC. I designed a few of them into some equipment at my old employer, where they ran using 65VDC @ 5 amperes without any issues at all. Gecko drive also makes a Step/Direction style drive for use with a DC servo motor that incorporates encoder feedback. The servo style Gecko drives are rated for 80VDC @ 20 amperes. I purchased one for myself a few years ago to play with. They work quite well. The issue with the Gecko drives is that their pricing starts at about $120.00 each - without the motor. Their not for the minimal cost design, that's for sure.

Last fall I designed a control system for my new Sherline table-top mill using the Gecko stepper drives. Though expensive, they do a great job and are very reliable - especially at 24VDC. The completed control cost nearly $1,000.00, as opposed to $1,500.00 for the mill and an additional $300.00 for the stepper motors.

I have accumulated another four Gecko stepper motor drives and will be building another control (this time a 4 axes) when time permits. The current control will move to the MAX-NC 5 mill and, the new 4 axis control will go to the Sherline mill with a rotory axis.

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

So to not short circuit the bridge, I'll need some sort of delay line. Would an inverter (or a few connected serialy) do the job?

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

No.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I thought I could build a pwm controller with dead band by having 2 comparators driving the gates on a 3 resistor chain driven from the triangle wave generator. The top comparator would always turn on after the bottom as the wave was rising. If the middle R was a trimmer, you could adjust the deadband. (Weren't they from San Francisco?)

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

daqq wrote:
So to not short circuit the bridge, I'll need some sort of delay line. Would an inverter (or a few connected serialy) do the job?

No! You need to ensure that the two FETs in the same half of the bridge are never, NEVER, EVER on at the same time.

For say, the left side of the H-Bridge, where the bottom FET is on and the top FET is off, you do the following:

You turn off the bottom FET, wait some time for the FET to turn all the way off, as specified by the datasheet "Turn-Off" time. When that time has elapsed, you turn on the top FET.

Going the other way, starting with the top FET on and the bottom FET off:

Turn off the top FET, wait some specified time, for the top FET to turn all the way off, as specified by the datasheet "Turn-Off" time. Once this time interval has elapsed, turn on the bottom FET.

The same logic applies to the the right side of the H-Bridge, as well as a Half-Bridge.

If you don't follow this convention, the best case is that you will consume excessive power in the FETs, generating excessive heat in the FETS. The worst case senario will be angry, smoking FETs, burned PCB traces, possibly a destroyed motor, burned up interconnect wiring, destroyed power supply, an exploding battery, etc... The lesson to learn here is: Use properly sized fuses in your development circuitry!

And, while on the subject, you do not want to be without safety glasses or a face shield when working on power circutry. ALWAYS WEAR SAFETY GLASSES OR A FACE SHIELD WHEN WORKING ON PORWER CONTROL CIRCUITRY!!! I was working an an Allegro 3 ampere H-Bridge several years ago. The thing decided to let go - even after it had been working quite well for several days - crap flew every where! A piece made a direct hit into my glasses, damaging them. What would have happened to my eye if I were not wearing glasses?

You should now fully understand why a commercially made H-Bridge has its economic advantages. It will only take one catastrophy during the learning curve to pay for an untold number of commercially made H-Bridges.

My philosiphy is... Let those guys who know what they are doing take the high risk! I simply can't afford to...

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Why not using uln2803 and its positive counterpart to implement 8 channel bridges!!!You can parallel them to increase current sinking cap!!!And i think there are special shift registers with high current cap, but i cannot remember part numbers.

I love Digital
and you who involved in it!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ali -

This is a VERY old thread!

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net