Basic problem with transistors

[giftiger_wunsch]

download Eagle from Cadsoft...It will do pcb and schematic layouts...
The schematic editor is brill..

Thanks, will do.

 

[audioguru]

I draw and modify other schematics with Microsoft Paint program. I copy and paste bits and pieces of other schematics, many from datasheets. Straight lines are made with the SHIFT key down.
Here is one of my schematics made with MS Paint:

 

[marcbarker]

Isn't everything upside-down in Australia and in New Zealand and motors run backwards?

Maybe the Coriolis effect works applies to motors too?...

Transistors have a wide range of spec's. Some saturate much better than others even if they have the same part numbers and are made by the same manufacturer. Most Zetec transistors saturate very well.
Can you run as fast and as far as me?

Zetec e-line BC650(?) or something, was a revolutionary transistor in its time. What are 'good spec' transistors nowdays? Can you suggest a really cheap 0.9V Vgs(th) mosFET?

audioguru and marcbarker <..>let's all just get along

It wasn't me it was him! ---->>>>>

 

[audioguru]

The Vgs(th) voltage rating of a Mosfet is when it is barely turned on. I don't care about it since I care about the Vgs voltage when it is turned on hard. It is a wide range of voltages for each Mosfet part number. Most ordinary Mosfets are turned on hard when their Vgs is 10V. Most "logic level" Mosfets are turned on hard when their Vgs is 4.5V.

The prices for electronic parts are so cheap that I don't look at them.

 

[Hero999]

The internal diodes on MOSFETs tend to be too slow to offer any real protection.

Whatever you do, don't make both inputs to this circuit high at the same time or Tr3 to Tr6 will all turn on and blow up.

It is possible to protect it from this by adding a Schottky diode to cut-off one of the inputs when the other is high. A Schottky is required because its turn on voltage plus the saturation voltage of Tr1 must be less than the turn on voltage of Tr2.

 

[indulis]

Body diodes are fast enough for SMPS secondary’s before self driven synchronous rectifiers become active... should work here.

 

[marcbarker]

I care about the Vgs voltage when it is turned on hard.

Good to hear that! I'm using this here cheap Fet to drive a 0.5 A load.

What gate Vgs voltage will I need for a nice low 100 milliohm Rds(on)?

**broken link removed**

edit: well according to this datasheet it's typically nothing like "4.5 Volts"!! More Like 2 V

 

[marcbarker]

Whatever you do, don't make both inputs to this circuit high at the same time or Tr3 to Tr6 will all turn on and blow up.

This might happen during uC start up condition. Or software crash.

Schottky diode i.e. BAT85 is great idea.

If it's still a concern, a low-dropout current limit circuit can be added to limit the current to 1 A. However, a series switch & current sense is going to be a bit tricky on 3 V, a simple fuse or PTC device (like used in cellphone battery) would probably be better.

 

[giftiger_wunsch]

Whatever you do, don't make both inputs to this circuit high at the same time or Tr3 to Tr6 will all turn on and blow up.

This is precisely what will be avoided using the green portion of my previous schematic. Using PA0 as a ground for PA1 and vice verca, no current will flow when both are high, since there will be no voltage between the two. See my earlier explanation Looking at the schematic you provided, I should be able to incorporate this aspect of my schematic itno yours without losing functionality. Once I've figured out how to use eagle properly I'll post an updated schematic illustrating how I plan to modify yours

I draw and modify other schematics with Microsoft Paint program. I copy and paste bits and pieces of other schematics, many from datasheets. Straight lines are made with the SHIFT key down.

I was doing the same thing with Microsoft Powerpoint, and then screenshotting the schematics: all of my schematics thus far posted have been created in this way. But it's becoming a bit tedious now so I'm hoping to find a decent schematic-drawing application which will make it simpler. Eagle seems easier except that its component libraries are hell to navigate...

 

[marcbarker]

This the green portion of my previous schematic. Using PA0 as a ground for PA1 and vice verca, no current will flow when both are high, since there will be no voltage between the two.

["TheDoneThing" cliché mode = ON]
You mean omit the ground connection between the uC ground and the 3V motor battery ground don't you? It's not the 'done thing'. Yes it's clever, but sometimes too clever in design isn't good !
[/OFF]

In this case it's because when you talk about the design with anyone, even if it works fine, it looks like the circuit's from mars and when people can't relate well to it there's a danger that it's not supported. Or misunderstood. As well as that you have problems when you're testing it, like "where do you connect the scope probe ground clip to" and things like that.

Apart from that, possible problem I notice is what happens when more than one motor is energised?

The circuit as drawn is meant to be linked to your uC's ground, but if you're sure it's going to work, please share your thoughts, ideally mark up the schematic with current flows.

 

[giftiger_wunsch]

["TheDoneThing" cliché mode = ON]

The circuit as drawn is meant to be linked to your uC's ground, but if you're sure it's going to work, please share your thoughts, ideally mark up the schematic with current flows.

I will do; you do of course have a point when it comes to what will happen when multiple motors are on; I will consider that when I come to it. I'm sure it won't be too difficult to separate that out.

I'm not really sure why it's so the un-done thing p) since as far as I can see it adds a pretty handy function, but as far as discussing it with others is concerned, I'll perhaps just have to give it a bit more explanation.

Either way I'll post my version of schematic here, complete with whatever explanations are necessary to make it understandable, and ask for opinions before I finally implement this idea.

 

[marcbarker]

I'm not really sure why it's so the un-done thing p) since as far as I can see it adds a pretty handy function, but as far as discussing it with others is concerned, I'll perhaps just have to give it a bit more explanation.

Either way I'll post my version of schematic here, complete with whatever explanations are necessary to make it understandable, and ask for opinions before I finally implement this idea.

Yes! It's the nature of science, is that if something is significantly different from the norm, for it to be creditable it'll come with sufficient explanation (with objective proof) to sell it to the critics!

 

[Nigel Goodwin]

What is the purpose of the protection diodes? I know that a relay coil produces a high-voltage spike which requires a protection diode to protect ICs driving them, but didn't realise this also applies to BJTs?

An IC is just a load of BJT's or FET's in a common casing, ALL of them really require protection diodes.

Here's another of my websites, with the circuit for an H-bridge used on a magazine robot - notice the motor is connected to the collectors for the reasons I mentioned previously. No diodes either, but a capacitor, which is reverse biased when going backwards - never seemed to cause any problems though.

 

[giftiger_wunsch]

I realise that an IC is simply a load of other components in circuits in a common casing, but it was my understanding that the protection diode was usually to protect these components from the high voltage spike which is produced by a relay coil or inducer; I didn't realise this could be produced by a BJT.

As for the cybot - brings back memories. I collected that magazine and built the cybot when I was about 10.

 

[marcbarker]

No one else seemed to pick this up, apart from saying 'circuit won't work', 'emitter drive is bad', 'fet won't turn on hard enough' etc.

.... Fall over with amazement!!... the circuit actually works*!. Not only that but almost exactly **broken link removed**too! This is the circuit from the OP, slightly modified.

The motor Battery is 2x D-cell NiCads. The motor gets 2.16 V of this, with a slight loss because of the transistors. Motor current is 198 mA. Base current is 1.2 mA.

The upper transistor is probably neither common-collector OR common emitter configuration.

Well I'm speechless!

**broken link removed**
*According to LTSpice simulation software

HERE IS IS THE LTC SIM .ASC FILE ATTACHED

 

[giftiger_wunsch]

What program did you use to simulate this? That looks extremely useful. Ergo probably extremely expensive.

Also, what is the function of the Zener Diode D1? Could you possibly modify your circuit to test my ground idea? That is, connect both PA1 and PA2 to the battery ground via signal diodes. If it's not too much trouble, of course.

 

[audioguru]

The BC337 transistor is spec'd to saturate well with a base current that is 1/10th the collector current. With a base current that is only 1.5mA and a collector current that is 150mA the typical saturation voltage is close to 1V (not 0.2V) and could be worse.
Then the motor will get not much voltage.

If you increase the base current to 8mA (the max allowed output current of the micro) then the output voltage of the micro will be much lower than 3.3V. Then the motor will get not much voltage.

 

[marcbarker]

What program did you use to simulate this? That looks extremely useful. Ergo probably extremely expensive.

Also, what is the function of the Zener Diode D1? Could you possibly modify your circuit to test my ground idea? That is, connect both PA1 and PA2 to the battery ground via signal diodes. If it's not too much trouble, of course.

get it from Linear Technology - LTspice IV Downloads and Updates

D1 you taught me it!, you should know It's that diode you put in, for the 'ground return', to do with the PA1/PA2 thingy so that you don't have a ground link between the uC ground and Motor Battery ground.

 

[giftiger_wunsch]

get it from Linear Technology - LTspice IV Downloads and Updates

D1 you taught me it!, you should know It's that diode you put in, for the 'ground return', to do with the PA1/PA2 thingy so that you don't have a ground link between the uC ground and Motor Battery ground.

Yeah, just a little confusing because some of the symbols are a little odd on the LTSpice schematic and PA0 / PA1 have an immutable voltage. I'm experimenting with LTSpice at the moment. Why did you choose a Zener diode rather than a signal diode though?

 

[marcbarker]

It's not a zener diode, it's a schottkey

The BC337 transistor is spec'd to saturate well with a base current that is 1/10th the collector current. With a base current that is only 1.5mA and a collector current that is 150mA the typical saturation voltage is close to 1V (not 0.2V) and could be worse.
Then the motor will get not much voltage.

If you increase the base current to 8mA (the max allowed output current of the micro) then the output voltage of the micro will be much lower than 3.3V. Then the motor will get not much voltage.

It's not possible easily to increase the base current to 8 mA, because the circuit only takes what base current it needs. It's a weird circuit configuration, it took me a while to figure Giftiger's circuit.

The closest analogy I can think of his circuit is an electronic version of a traditional hydraulic car power-assist steering system. The Motor battery is the hydraulic pump, the transistors are the pinion and control valves, and the uC the driver and steering wheel. There's even a 'force-feedback' effect going on: as the motor current increases, so does (in proportion) the output current from the uC.

Hey Giftiger mate? Maybe you could patent this? Serious! Or at least print out a few copies and send them to yourself sealed Registered Mail and keep the unopened letters safe!