MINI PCB/Engraving Mill PSU - Schematic Help

[unicorn13]

Hi All

It would be appreciated if some one with more experience could have a look at the attached schematic.

It is for a Desktop machine power supply, Component values are known for the Soft start and +5v regulated but components for the stepper power are still not defined, toroid size, voltage, Amps etc.

I expect that around 36-42V output min 5 to max 10 amps possibly, as I need to work out which steppers I will use, there will be three (max 2A peak each?) but not all drawing full current simultaneously.

Machine dimensions Will end up 18"X18"X18" approx, It will be a moving gantry type constructed from Aluminium.

Mainly for Light engraving, PCB Routing/Drilling etc.

comments, corrections and guidance appreciated.

To all here many thanks and a happy New Year.

Andy

[pminmo]

A couple of questions. Is there a reason TR2 isn't switched? D1 is a gonner as soon as you plug it in as the is no current limiting for it. If your looking at 10A 36VDC, I'm not sure why you need a soft start circuit, and if you do, it would be much easier to switch the DC side to disconnect the Caps. Is there some specific reason to add the complexity?

[Al_The_Man]

As Pimino said, for that VA I wouldn't think you will need softstart, where is the delay if K2 is on, on switch on, unless you are looking at the relay pick-up delay.
Why is the stepper power supply referenced to ground through the caps?
Al.

[unicorn13]

Hi pminmo & al_the_man,

Thanks for replies so far, I'm still new to cnc so trying to find best solutions, the schematic is built from several others, mostly a learning curve.

1. TR2 would be switched but just not shown, forgot to add it

2. The original soft start circuit was linked to main live via resistors and caps, to drop the voltage (I DON'T LIKE THAT IDEA) so in attempting to use tr2 would inevitably highlight any mistakes in trying to convert the circuit.

3. K1 is there to take any of the TR1 switch on issues if they arose (according to original circuit) away from the front panel on switch.

4. according to circuit description sequence is power on K1 energizes allowing power flow through resistor bank on K2. 100ms or so later K2 energizes taking resistor bank out of circuit. Although not said in original descriptions I believe that K2 is never energised on initial power up, so K2 would not be in an 'on' state initially, or have I misinterpreted the schematic?

5. Stepper supply circuit was taken from a description and image of working Class-A amplifier supply I used it as was shown because it would give the volts/Amps output needed, circuit described as a Full Wave CT Linear supply (wrong choice of supply?).

Idea of using softstart was because of reading a lot of message regarding inrush limiting and blown fuses. So I thought better safe than sorry!

And of course +5v to supply controller cards. I know that works as I built it about 15yrs ago.

The plans for the machine have not arrived with me yet, so nothing is set in stone, but here's the url of the machine (all in German):

http://www.dh7pm.de/html/page_01.htm


All this, is planning and reworking to get the best and safest ideas before building, after all there are a lot more threads to read yet.

Heck I've still go to work out what stepper motors (they look like a NEMA23 type), controller and interface to use, was maybe thinking PIC based seeing I have a PIC programmer or just straight printer port!

Many thanks
Andy

[CJL5585]

Andy,

Keep the main fuse and switch. Delete everything from the switch up to the transformer primary. Connect the Hot 120 VAC from the switch to the primary of the transformer. Connect the neutral to the other side of the primary. Connect the ground wire to power supply chassis. Wire up the rectifier, capacitors and bleed resistor.

Delete everything relating to soft start. Soft start is only required for circuits with very high inrush currents.

Keep the 5 VDC regulator circuit. Get a small 6 volt or so transformer to supply the regulator circuit voltage.

Build everything as simple as possible. Simple is better, especially when one has to troubleshoot and repair something he has built.

Good luck.

[Al_The_Man]

I would either leave TR1 sec. C.T. unconnected and either ground the DC common or leave as isolated supply as per your preference, personnally I ground all power supplies common to a ground plate.
It depends on the what the supplier of your drives recommend and if you prefer to conform to any particular method.
Al.

[CJL5585]

My previous statement:
Connect the ground wire to power supply chassis. Wire up the rectifier, capacitors and bleed resistor.
---------------------------------------
I should have stated connect the Utility company ground to power supply chassis.

I also agree that the transformer secondary center-tap should remain un-connected.

[unicorn13]

Hi again

I have been thinking about what has been said, How about this as an alternative, see attached.

Also found a link on EDN regarding Bleed Resistors and the proposition they had in the design for a less wateful way of dumping cap voltage and running considerations of the bleed resistor constantly in circuit while running.

It's a lot simpler as CJL5585, Pminmo and Al_The_Man suggested.

What Did any of you chaps think of the CNCIII machine plans, I believe it should do all that I want.

Regards
Andy

[Al_The_Man]

I have seen this method done before, personally I would not use it, just because if ever the relay contact stuck or welded together, it puts the primary potential on the secondary supply, maybe a remote possibility, but I don't think it is good practice.
If you had a DC common that was connected to Ground plate, it would put ac on to the caps.
Al.

[unicorn13]

Hi Al

Didn't see that, how would adding a 1k 5w WW resistor after C4 in last schematic, and removing R1 and R2.

Also is it just 1 resistor needed or 1 per cap at lower values?

Regards
Andy

[Al_The_Man]

You only need one bleed-off resistor, 1k will give you 33ma for about a 18sec bleed-off time.
Al.

[pminmo]

There is one aspect that bothers me, but it's done, and that's switching the neutral with the HI side AC. If the Hi side only is switched and the switch fails closed, you will still have power so the operator knows something has failed. When switching both sides, if the high side fails but the low side doesn't, DC goes off, BUT anything tied to the low side not has the high side potential. It's the same situation as to why NEC requires the high side switched by code.