Hopefully a SMPS...

Thread Starter

Allenph

Joined May 27, 2015
76
Linear is way more forgiving of current and power spikes than SMPS based systems are. As far as heat goes it's only a problem if you designed the circuit wrong and used way too small of rated devices and heat sinks.

If the primary load is almost all motor driver power then the voltage and current filtering is even less of an issue and a simple transformer to bridge rectifier to good old heavy capacitor bank is more than sufficient and clean enough and the dedicated more sensitive logic circuits can be ran off their own independant regulated power supply so they are isolated and for the most part immune to motor driver related noise.
But then I'm going to end up building two power supplies in one instead of one which the 3D printer module just splits into two separate power channels...right?
 

shortbus

Joined Sep 30, 2009
10,045
I was just under the impression that using a linear supply like you suggested wouldn't work well because there would be some near constant current spikes for hours on end...isn't that a fire waiting to happen?
I don't know about what they are currently advising but I have 5 or six different commercial stepper driver modules and the manuals for all of the suggest linear power supplies. And specifically recommend not using a switcher type power supply. So what ever stepper driver module you will be using, you might want to check what they need before spending time and money on this. But in the end it's your call.
 

Thread Starter

Allenph

Joined May 27, 2015
76
I don't know about what they are currently advising but I have 5 or six different commercial stepper driver modules and the manuals for all of the suggest linear power supplies. And specifically recommend not using a switcher type power supply. So what ever stepper driver module you will be using, you might want to check what they need before spending time and money on this. But in the end it's your call.
I'm not sure why this would be so...with 3 stepper motors and a possible CNC DC motor each with relatively high torque you could end up with some serious voltage drops in a linear power supply. Plus I would rather not make two power supplies...one for the control electronics and one for the motors.

I'm not using modules. The control electronics are all on one PCB and it's a custom made PCB. The steppers all run on home-made h-bridges with flywheel diodes to prevent feedback. The motors and the device is controlled by a microcontroller which interprets signals from a TI wifi IC and runs the stepper motors. All the actual processing is done server side instead of client side.
 

shortbus

Joined Sep 30, 2009
10,045
Like I said your call. And I suppose your going to run the steppers at the voltage printed on them too? You better do some more research. Your DIY drivers will never make some one who knows think you bought this machine. Good luck. http://www.geckodrive.com/
 

Alec_t

Joined Sep 17, 2013
14,280
There are still probs with your circuit (so far as I can make it out, it being so small):-
The bridge rectifier seems incorrectly oriented.
Both M1 and M2 seem to have source and drain reversed.
How do you get M2 gate several volts above its source to turn it on?
 

tcmtech

Joined Nov 4, 2013
2,867
But then I'm going to end up building two power supplies in one instead of one which the 3D printer module just splits into two separate power channels...right?
As with my brothers rather sizable three axis CNC router table I know that running two independent power supplies is an almost absolute necessary and that as others stated SMPS don't work worth crap on stepper motor drive systems.

As for secondary power supply power I am guess your systems runs off of typical 5 VDC power which can easily be supplied by most any common USB type power adapter which makes the secondary power supply building a non issue in my view being you can walk into most any electronics store that has a recycling bin for your old power packs and other old portable electronics and grab a few for free.

I'm not sure why this would be so...with 3 stepper motors and a possible CNC DC motor each with relatively high torque you could end up with some serious voltage drops in a linear power supply. Plus I would rather not make two power supplies...one for the control electronics and one for the motors.
I do not follow your reasoning being that it's well known that a properly designed SMSP that is rated for say 4 amps has very little reserve power capacity beyond its 4 amp rating before it totally shuts down whereas any half decent linear unit can take substantial overload without problems unless it has a fold back current limiting system with a very sharp current overload knee point which a home built one would not have that problem unless you added it on purpose.

As others have pointed out your circuit is a poor choice for what you are wanting to power with it. Especially so if you plan on trying to run multiple stepper motors and your logic systems off it at the same time. :(
 

Thread Starter

Allenph

Joined May 27, 2015
76
There are still probs with your circuit (so far as I can make it out, it being so small):-
The bridge rectifier seems incorrectly oriented.
Both M1 and M2 seem to have source and drain reversed.
How do you get M2 gate several volts above its source to turn it on?
I noticed the bridge rectifier issue. I corrected it in an updated diagram but haven't posted it yet. I also noticed the the zener diode has no DC path to ground, so there never will be an input on the gate. I'll be changing the position of R13 to accommodate that.

I don't think they're reversed?

What do you mean by "How do you get M2 gate several volts above its source to turn it on?"
 

Thread Starter

Allenph

Joined May 27, 2015
76
As with my brothers rather sizable three axis CNC router table I know that running two independent power supplies is an almost absolute necessary and that as others stated SMPS don't work worth crap on stepper motor drive systems.

As for secondary power supply power I am guess your systems runs off of typical 5 VDC power which can easily be supplied by most any common USB type power adapter which makes the secondary power supply building a non issue in my view being you can walk into most any electronics store that has a recycling bin for your old power packs and other old portable electronics and grab a few for free.



I do not follow your reasoning being that it's well known that a properly designed SMSP that is rated for say 4 amps has very little reserve power capacity beyond its 4 amp rating before it totally shuts down whereas any half decent linear unit can take substantial overload without problems unless it has a fold back current limiting system with a very sharp current overload knee point which a home built one would not have that problem unless you added it on purpose.

As others have pointed out your circuit is a poor choice for what you are wanting to power with it. Especially so if you plan on trying to run multiple stepper motors and your logic systems off it at the same time. :(
All right. I'm a believer now. I think I've come up with a way to do both power supplies on the same board.

What if I use a transformer that outputs 12V, one branch of the power supply is my SMPS that takes the 12V down to 5V, and another branch is just a huge bank of capacitors that supplies the motors.

Regardless, I don't want to use an off the shelf PSU, even for the 5V. Not trying to be a pain, it's just how I want it to work. xD

So, this power supply actually will help me, and is applicable.

A couple questions:

1) How does one figure out the amount of capacitance required for motor loads in a linear power supply?
2) After switching the bridge rectifier around, and grounding the other side of the zener is there anything else wrong?

(I'll post a new diagram with the updated specs momentarily.)
 

Thread Starter

Allenph

Joined May 27, 2015
76


Okay, so you were right. @Alec_t, I switched the MOSFETs around. I also gave the zener a path to ground. In addition, I also switched the inductor and diode...so now It's actually a buck converter.

The last problem I've heard mentioned is that I need a voltage higher than the voltage on the source of M2 to turn it on...why would this be?

Any other issues?

Thanks for all the help guys. I really appreciate it.
 

tcmtech

Joined Nov 4, 2013
2,867
All right. I'm a believer now. I think I've come up with a way to do both power supplies on the same board.

What if I use a transformer that outputs 12V, one branch of the power supply is my SMPS that takes the 12V down to 5V, and another branch is just a huge bank of capacitors that supplies the motors.

Regardless, I don't want to use an off the shelf PSU, even for the 5V. Not trying to be a pain, it's just how I want it to work. xD
That should work just fine running the motors off the unregulated DC power and having the more sensitive stuff running off its own 5 VDC switching type sub power supply.
 

Thread Starter

Allenph

Joined May 27, 2015
76

Okay. I figured I was close enough to see some kind of positive result in LT Spice. I was very wrong.

A few things to notice:

1) M1 is needs to be a DEPLETION mode P-Channel Mosfet. This is REALLY important because I'm essentially doing the exact opposite of what I want to. The function of M1 is to INCREASE in resistance from zero based on the zener feedback channel. Thus reducing duty cycle, and therefore reducing voltage.
2) The resistor connected the zener feedback channel is now AFTER the zener. This regulates the current we're pulling off of the load, AND gives us a path to ground without making the gate of M1 electrically common to ground.
3) I gave the input to M2 a resistor in a similar manner to M1. Otherwise, there would be no voltage at the gate because the gate is isolated from the rest of the transistor.

Some things I'm unsure about:

1) I can't find a depletion mode MOSFET for either NPN nor PNP type MOSFETS in LT Spice.
2) I don't know what value to use for M2's gate resistor.
 

Alec_t

Joined Sep 17, 2013
14,280
The last problem I've heard mentioned is that I need a voltage higher than the voltage on the source of M2 to turn it on...why would this be?
Because that's the way enhancement mode N-channel MOSFETS work! Unless it's specifically a 'logic-level' type, the FET will generally need a Vgs of at least 10V to turn it on fully. If it doesn't turn on fully it will run HOT.
 

shortbus

Joined Sep 30, 2009
10,045
You really should read what is said on site the link I gave has to say, even if doing it on your own. Unless your DIY stepper driver has more than a couple of H-bridges, you are going to be very disappointed in your results. There is much more involved in driving a stepper motor in a CNC than just making it move. Using more than the motors rated voltage is just one of them, but an important one.

A proper motor driver is more then making the correct digital codes to work the H-bridge. Do a search on the site for DIY stepper drivers and see the number of people that come asking, "why doesn't my driver work?" With the large number of ready made driver chips out there, why beat your head against a wall to do your own. Unless you have unlimited time and money to devote to this project. Even the commercial sellers of CNC printers don't develop their own motor drivers, power supplies, etc. Your not building your own stepper motors are you? Or extruding you own aluminum for the frame. :)
 

tcmtech

Joined Nov 4, 2013
2,867
Similar thoughts here as well being every aspect of making a functional CNC machine of any sort is now pretty much cheap easy to to work with plug and play type primary components and it's just a matter of matching one composite rating to the next one that it has to work with.

Your power supply issues being one of them. I for one cannot follow the logic and reasoning behind your present circuit and the need to try and make it work as a basic SMPS system given that for what you are doing there are many far simpler and far more reliable designs out there. Most that can be built from one cheap chip and half a dozen extra components at most.
Your oscillator circuit alone has more parts and cost in it yet at this point does not work let alone has any working interaction to run any form of buck converter with any degree of output regulation.

As for the learning experience everyone here can appreciate the self taught hands on approach to a point but for most people there comes a point of realizing that what they want and how to achieve it has to be reasoned out to how it will meet the end goal of having a working and workable device and what it will cost in terms of time labor and financial outlay.
 

tcmtech

Joined Nov 4, 2013
2,867
My brother went though the same mentality when building his first version of his CNC. Loads of time, expectations of ridiculous tolerances, one off designs for almost every part of the system and in the end it was a huge waste of time money materials and for the most part a dismally sloppy contraption that didn't work worth crap.

Now version 2.0 was well built mechanically and electrically using correctly planned out and realistic designs made entirely out of off the shelf plug and play components that have ran beautifully day in and day out for a few years now with little maintenance or fuss. :cool:
 
Top