My Schematics are strange...?

 

My Schematics are strange...?

Not stranger than you Ok let me work on some more insults (ups did I say that) No I mean circuit ideas. As a start I think it would be better to place a schmitt trigger inverter as a input buffer before the Optocouplers. And then use the Optocouplers in a inverting setup. Else you will have little or less control over the LED current for the Optocouplers. Using no pullups for the dip switch may be OK. Since the input stages on the TB6550 do have internal pull down resistors equal to about 100K. But using pull ups resistors will make the design more rugged for against noise I will think. I see if I can make something (schematic) tomorrow. Is the purpose of this board testing only. Or is the plan that the board should evolve into something that can control one axis of a CNC machine. Controlled by PC printer port

 

See...now that's more like it

This circuit is for me to test my Nema23 motors properly.
The actual CNC axis drives will be similar ...

I need to test the power dissipation of the TB chip and also how I can miniaturize the complete axis control PCB's into a computer unit...

So to say this PCB is to study

{ed}
I think I should use a PIC to do the start up sequence of the TB.
Am I right ?
This way it will be safer for the chip, isn't it ?

 

I have done some doodleing. And I came up with this. As a suggestion. It is draft only, decoupling caps are missing sevral places

 

Nice, but I have a couple of questions. Please bear with me, I'm trying to learn and understand.

1. What is K1 and K2 doing?
2. I thought PIN9 on U3 had to be connected to positive.
3. If P1 PIN4 is LOW, then U3 PIN 1&2 are LOW, are U3 PIN 15 & 16 also LOW, and the LED will turn on? (I thought PIN 9 would have been connected to +5v and the LEDs would go to ground)
4. I guess J1 is +24V power plug. But where is +5v power plug or regulator?
5. Why SGND and GND? Why not only GND?
6. Is SW1 the reset switch?

Hope I'm not to ignorant to cure...

 

Nice, but I have a couple of questions. Please bear with me, I'm trying to learn and understand.

1. What is K1 and K2 doing?
2. I thought PIN9 on U3 had to be connected to positive.
3. If P1 PIN4 is LOW, then U3 PIN 1&2 are LOW, are U3 PIN 15 & 16 also LOW, and the LED will turn on? (I thought PIN 9 would have been connected to +5v and the LEDs would go to ground)
4. I guess J1 is +24V power plug. But where is +5v power plug or regulator?
5. Why SGND and GND? Why not only GND?
6. Is SW1 the reset switch?

Hope I'm not to ignorant to cure...

Hope I'm not to ignorant to cure...

No not at all this is a discussion forum. And my "design draft" is not in any way carved in stone.

1. K1 and K2 is doing nothing at the moment. They should have been removed. The TB6550 may be fried if not a start up sequence is followed. This is mentioned in the data sheet at page 28. R!f@@ did mention using a PIC to ensure this start up sequency is followed. And then the relays would have been used. I was just ahead of my time.
2. You can do it but then you will have a a diode voltage drop. Take a look here http://www.ti.com/lit/ds/symlink/uln2003a.pdf and figure 20 for a practical use of pin 9
3. In this setup you can simplify. By think as if you have a high on the input. The output will be grounded. This is note quite true. As you will have the ULN2003a COLLECTOR-EMITTER SATURATION VOLTAGE drop. For LED about 20mA. I think around 0.7 volt se figure 11 in the datasheet.
4. Not included yet. Good thinking to see that.
5. It is a trick to avoid noise from the motors. You want to have one dirty "track" for the motors and one clean ground for the logic. The grounds will then meet up as close to the ground on the power supply as possible. using separate wires as one example.
6. This has to do with the start up sequence mention before. And perhaps not a good design at all. R2 is misplaced. As for the moment think of it as gone. Then the switch is open. The output of U1 will be tri stated. Internal pulldown resistors in the TB6550. Will then pull the control signals LOW. And this is needed during the start up sequence. Later this may be controlled by a PIC. And using the setup will make more sense

 

I have made some updates to the schematic. As for now I not very happy with "grounds for grounding" scheme. Much more work will be needed on this topic.
Edit: Note this schematic is also only a draft.

 

Rifaa,
I just took a look at your Eagle schematic you attached in #114.

I am afraid that you don't understand when to use wires, and when to use a bus.

Wires connect pins to pins. If you have a whole bunch of wires going from one place to another, you can stop the wires at a bus - BUT! You need to NAME and LABEL the wires where they enter and exit the bus, or Eagle won't know what runs to what!

You have tried to use wire as a bus. As a result, many pins were all shorted together.

You tried to use bus as wire. That doesn't work; you need to use wire between pins.

You also need to use junctions where more than one wire connects to a pin, or when two wires connect to each other.

You cannot place a junction on top of a junction. If you need to add another wire to a pin, then delete the existing junction, add the wire, then place a junction on top. You can't have more than 1 junction in one place.

You need to use ERC early, and use it often. If you had tried to create a PCB from that schematic, you would have a big mess.

[ETA]
Here, I've fixed up some of the stuff, got rid of some of the labels that weren't used like VDD and 5v (you still have some NAMEs that use that). Have a look at where I corrected the bus down below, and do the same to the upper right corner bus.

Also, there's some "leftovers" - the two transistors, you had a wire connected from somewhere above it, I put a "Fix this Rifaa!" note there. If it's supposed to go in the bus, you need to give it a name and connect it to something on the other end.

 

I have made some updates to the schematic. As for now I not very happy with "grounds for grounding" scheme. Much more work will be needed on this topic.

How about those decoupling caps...

Rifaa,
I just took a look at your Eagle schematic you attached in #114.

I am afraid that you don't understand when to use wires, and when to use a bus.

Wires connect pins to pins. If you have a whole bunch of wires going from one place to another, you can stop the wires at a bus - BUT! You need to NAME and LABEL the wires where they enter and exit the bus, or Eagle won't know what runs to what!

You have tried to use wire as a bus. As a result, many pins were all shorted together.

You tried to use bus as wire. That doesn't work; you need to use wire between pins.

You also need to use junctions where more than one wire connects to a pin, or when two wires connect to each other.

You cannot place a junction on top of a junction. If you need to add another wire to a pin, then delete the existing junction, add the wire, then place a junction on top. You can't have more than 1 junction in one place.

You need to use ERC early, and use it often. If you had tried to create a PCB from that schematic, you would have a big mess.

[ETA]
Here, I've fixed up some of the stuff, got rid of some of the labels that weren't used like VDD and 5v (you still have some NAMEs that use that). Have a look at where I corrected the bus down below, and do the same to the upper right corner bus.

Also, there's some "leftovers" - the two transistors, you had a wire connected from somewhere above it, I put a "Fix this Rifaa!" note there. If it's supposed to go in the bus, you need to give it a name and connect it to something on the other end.

Darn. Don't have Eagle installed on my computer at work....

 

Sorry nerdegutta, I didn't know.

I moved things a bit so the unconnected wires going to the bus are showing.

Rifaa needs to fix all of that stuff himself, as the way he wired it up originally, it couldn't be told exactly where it was supposed to go.

You made some assumptions that I am certain are not correct; for example the collectors of the transistors, and the base of the left transistor (Q2). I am not sure what was supposed to be connected to which transistor.

He has the RESET line connected to some stuff, but where that goes I'm not sure.

I just noticed I left N$10's label floating around. As soon as you try to move it, it will show you that it belongs to the wire below it.

I eliminated some of the power traces, replacing them with symbols - but Rifaa originally had 12v going to the 74 series ICs - that's a no-no - and also has the TB6560 powered with 12v. I just put a note there, "12V? Are you SURE?"

 

You made some assumptions that I am certain are not correct; for example the collectors of the transistors, and the base of the left transistor (Q2). I am not sure what was supposed to be connected to which transistor.

Yes I did. Got really confused by the mixing of bus and trace. And the transistor connections...

Thanks.

 

How about those decoupling caps...

I can add them later, if needed. We need some feedback from the OP at this stage

 

Yes, I guess we do. SgtWookies post #81, have an attachment. In that att there is a fuse on the TB6560's Vma and Vmb. Something that should be considered?

 

Yes, I guess we do. SgtWookies post #81, have an attachment. In that att there is a fuse on the TB6560's Vma and Vmb. Something that should be considered?

I would use a fuse, but not place it on this board. But closer to the power supply

 

Hey Guys.. been bzy..I'll give feedback later tonight.

The Hospital Dryer Started acting up.

Later I will edit this post.

{ed}

First I like to tell you that the schematic was just something I threw up referring to something long ago. Then I believe I have no way of doing a PCB as I recall
The diagram was not intended to a transfer and was a rough idea and I needed something to refer to. I did made that circuit on a vero but actually did not fire it up, I thought it would be wise to do it on a PCB properly.

@ Sgt.
Thanks for the effort. It's not that I did not know how to use the wires and bus, it's was just a sketch. I believe you should realize that I did some boards and even made some libraries. This particular one was not close to a finish.

But the changes you made were quite helpful though.

To tell u the truth I also had doubts about that circuit cause it was something I found while googling. I knew there were errors but heck I dunno why I never bothered to show it and discuss it before. maybe I was waiting for the right time and things.

Anyways it was for the better.

@t06
Your diagram has a 2003 which I believe I have lying around. But Why did you use a another CMOS.. Why the 74HC one.

@ Jens
You are going too fast on this for me....I have too much work on hand to do any of my stuffs.

Still like to thank you all for everything.
Things are looking good and I trust that I don't have to worry about any of my circuits since you guys are here.

I am looking into the amends you guys made ..have all of em printed and studying it.

I believe I will be adding a PIC for the start up sequence.

Sgt and Jens, you guys can relax for now. I will inform you once I draw the final one.
But before that I like to know what t06 says about his diagram.

 

The diagram was not intended to a transhfer and was a rough ide

Everything starts with an idea!

 

. @t06 Your diagram has a 2003 which I believe I have lying around. But Why did you use a another CMOS.. Why the 74HC one. Sgt and Jens, you guys can relax for now. I will inform you once I draw the final one. But before that I like to know what t06 says about his diagram.

The ULN2003A has a 2.7-kΩ series base resistor for each Darlington pair for operation directly with TTL or 5-V CMOS devices. I decided to use this circuit because it is cheap and common. The 74HCT series may drive 20 mA on a output, but in the specs the absolute max current draw pr circuit is 50 mA. In your circuit you had 3 LEDs. If all of them are turned on. You will need at least 60 mA at least. I know you want to drive your LEDs close to max current. You also needed some driver for the 3 optcouplers. And the same will also apply here. I think using the ULN2003 will give you a more robust circuit. compared to using a 74HCT06 as driver for the LEDs and the optocouplers. By the way do think the opto couplers are realy needed. Instead of using opto couplers why not run separate grounds wires for the GND and SGND signals all the way to the power supply.

 

I know you guys are set on using the TB6560 (big mistake IMO) but have a look at this. Since your going to use a uC for the start up, why not just use one for the whole thing? All the PCB files and the coding for the uC are already done and the files and code are open source/free. Or can be bought for probably less than the cost to actually do it. I just got one of the bare boards and pre-programed uC's, board is super nice. http://avrstmd.com/

And the circuit includes a 'test' function, to do what R!f@@ wants right now.