H-bridge

[AtomSoft]

OK i hear these are simple to make but seem to burn out transistors... you think these would do : IRF820

 

[Misterbenn]

depends on what your making .... i mean you could be using 5V @10mA or 1000V at 100A....

You need to give LOTS more detail than just 'will this be ok'

 

[AtomSoft]

its to control a motor from a CD ROM just want to make it for forward and backwards. Im using 3.3v to 5v

 

[Misterbenn]

I'll give you an answer fitting to the level of detail you've provided.

Probably yes.

 

[TechnoGilles]

For simple motor control with H-Bridge, I usually use the all-in-one L293D. Simple to use, directly from the MCU.

 

[AtomSoft]

For simple motor control with H-Bridge, I usually use the all-in-one L293D. Simple to use, directly from the MCU.

Thanks for the simple and useful answer. Ill order one. Seems simple to use

 

[Reloadron]

its to control a motor from a CD ROM just want to make it for forward and backwards. Im using 3.3v to 5v

Jason, isn't a CD ROM motor sort of like a 3 phase stepper motor. Thus, reversing isn't as simple as a plain H bridge.

Ron

 

[AtomSoft]

You would be right but im talking about the motor which controls the CD TRAY

 

[Reloadron]

You would be right but im talking about the motor which controls the CD TRAY

Why did I have that feeling after I posted.

Ron

 

[cr0sh]

You might want to find out what the current needs of the motor are, before you order the L293; if it needs more than 600 mA, verify with L293 you are getting (one allows for only 600 mA of current, but a different version goes up to around 1 A - note you will have to provide a heat sink). Another thing is that it is possible to stack L293s (I've only heard of doubling them up - I doubt you can go beyond that, and even doubling them up sounds iffy to me, since the outputs are transistors, not fets - but there's anecdotal evidence that it works). I doubt the motor will require it, but the L298 is the next level up (well, for these old-school h-bridges) for current rating; they'll take you up to around 4 A in bridged mode (there are two 2 A bridges on each IC), but you will need a heat sink, once again (multiwatt-15 heat sinks aren't easy to find, btw). Also, they require a special pad layout (it's not a standard 0.1 inch grid), so prototyping with them isn't easy (I've used this guy's adaptors with success, though: https://www.jrhackett.net/L298adapter.shtml). Another interesting option for low-current h-bridges are these small SIP devices (Rohm BA6886N): **broken link removed**

 

[AtomSoft]

I actual have some transistors i found. heh i will test them first they are

TIP42 - PNP 40v @ 10A
TIP3055 60v @ 10A

 

[AtomSoft]

You think this is good ?

 

[AtomSoft]

ok I tried making this but it doesnt work not sure why

but 12v

 

[Joe G]

i used an opticoupler from four pic pins. used a 330ohm res on the bases of 2n2222/2n2907 (with pulup/puldwn res ), i also used shotkey diodes from gnd=motorout=pos. the motor i used is from the same as your project , it powered by 3.6v and has a stall of about 100ma.

 

[Reloadron]

Jason, what are R1 and R2 for? What voltage are you applying to the bases (A&B)?

I have one of those little motors lying here so just for the hell of it I ran it off a 12 volt battery (12.7 volts). Free running under no load it only draws about 35 mA and in a dead stall it gets up to about 800 to 900 mA. If you have a 2N2222 (NPN) and 2n2907 (PNP) they would likely work and I would use maybe 1 K base resistors to start if you are applying about 5 volts to the bases.

What you have is substantial overkill for what these little motors need.

<EDIT> I see Joe G posted, you had a much lower stall current than I got but yeah, I agree on the transistors. </EDIT>

Ron

 

[debe]

How you have wired it it will create a short on the power rail. A should turn on T2 & T4. B should turn on T1 & T3.

 

[cr0sh]

ok I tried making this but it doesnt work not sure why

but 12v

Well - if you didn't burn out the transistors...

First off - you need base resistors on all the transistors - start with 1K resistors, that should be OK (there are ways to calculate what you need). You'll also want to add your diodes back in for back-EMF.

Now - you want to connect "A" to the upper-left PNP and the lower-right NPN, and "B" to the upper-right PNP and lower-left NPN. The way you have it set up, current won't flow thru the motor (and instead is shorting thru the transistors - which is BAD and called "shoot-thru") when activating A or B. You will also want to increase the voltage of the motor power source, as each transistor is dropping about a volt (leaving you with only 3 volts to go thru the motor - which is OK if the motor is a 3 volt motor, otherwise not).

Finally - depending on the current needs of the motor (not high likely) - you may need small heatsinks on the transistors (but I doubt it - those transistors are waaaay overkill for a small motor like you have)...

 

[Reloadron]

How you have wired it it will create a short on the power rail. A should turn on T2 & T4. B should turn on T1 & T3.

Whoops, I missed that....

Ron

 

[Joe G]

is connections A and B both high/low at the same time? that would make a differance as to the bases connections, as cr0sh stated i assumed the +/-

 

[debe]

Why not go looking old VCRs theres plenty of free chips that will do what you want.