Simple power supply question

[jseiler]

I have a PR32 and am about to upgrade the power supply to 24V. I bought a dual power supply, 24V and 5V. The question I have is...

Is there anything besides the motors that use the 12V in the standard configuration? If not, can I simply disconnect the standard supply and use the new supply to run the +5 for the logic and the 24V for the motors? The previous threads seemed to indicate there may be a need to keep the old supply connected, but I suspect that's only for the 5v logic. Is that correct?

John Seiler
-- ps. I have the 680 ohm current resistors and 3 100uf caps.

[Brady Watson]

John,
I am not sure about running 24v on the old PR boxes...you should probably call SB to find out if the motor drivers can handle the increase in voltage. The motors themselves, will be fine at that voltage.

Also, be sure that you are using a non-switching power supply with enough ampacity to move the motors.

-Brady

[jseiler]

Shopbot said this was fine with the resistor change and adding the caps to stiffen the PS on the motor drivers. I can't see anywhere where 12V is necessary except for motor drive, but I was posting to see if anyone else knew for sure before I toast something.

I'm not sure I understand the reason behind using a linear over a switcher if the current capacity is ok and there are stiffening caps on the rails for transient suppression. The original 12v supply is a switcher and only 9 or so amps.

regards,
John Seiler

[gerald_d]

The "modern" power supplies are also switchers.

[Brady Watson]

I overlooked the fact that you have the caps...linear supplies intrinsically have caps on them. Switchers don't always have them...

You are essentially concerned with 2 things in regards to the 12/24v. You want to make sure that you have the required power for the control board itself, and then power for the steppers that will be fed to the drivers. If you've come this far, I am sure you can figure it out. Why not just leave the original supply there to run the electronics and then add your hotter supply for just the motors???

-Brady

[dingwall]

What's the advantage of doubling the motor voltage?

[Brady Watson]

Speed.

[jseiler]

Sheldon... Its an inductance thing. Motor windings resist rapid changes in current. Running at a higher voltage allows one to come up to a steady state current more quickly. The other advantage is (all things being equal)....speed is proportional to voltage, torque is proportional to current. I should be able to get the same torque at twice the speed even at the same winding current.
I hope.

Brady-- I'd considered that, but since I bought this machine used and the included power supply is rather old and the new one has +5V out as well as 24, why not use only the new one, leaving the old one in the case as a backup in case of failure. That is...if the +12V is only necessary for motor drive. If that's not the case, I'll have no other option but to run both.

(Why a PR32? It was cheap, maintainable, small enough to fit into my garage in a full 6'x6' enclosure, runs off of 120V. Someday when I make enough money with it, I'll buy bigger, much bigger.)

John Seiler

[gerald_d]

It begs the question.....If running on 24V is obviously that good, why were the early SB's produced with only 12V?

[Brady Watson]

Gerald,
I think that Shopbot's mantra all along has been reliability, within the constraints of affordability. I imagine that at the time the early PRs were being made, power supplies suitable for a CNC were costly. If you saw one of the early PR power supplies (not sure if John's is same) you'd see just how beefy the thing is!

I am also fairly certain that the early PRs ran NEMA 17 size motors, unlike the later PR and PRT 34 size motors.

John,
Another option for your PR32 is to run a Xylotex board (will handle up to 2.5A per driver), a torid-style power supply and Mach2 or other controller interface. This would cost you a total of about $400, but you'll lose the ability to run SB software & need to learn G-code.

-Brady

[jseiler]

PC power supplies and cases are cheap???

I had emailed shopbot about doing this upgrade and someone there forwarded this to Ted, who responded. He seemed suprised that I had a 12V one because many had shipped with higher voltage controls.

John Seiler

[Brady Watson]

Then you must not have the PS I was referring to in my previous post...It sure wasn't a PC supply...which by the way, are NOT adequate or recommended for running a CNC. The current ratings on those supplies are not conducive to running 3 steppers simultaneously.

-Brady

[jseiler]

It really is just a PC power supply. I may have to consider running just the x and Y on the hotter supply and leave the z on the 12V one. That should help some. I guess if this works, I should start looking for a beefier supply. What would you recommend?

Looking at this thread:

http://www.talkshopbot.com/forum/messages/28/65.html

seems to indicate that the PC power supply is just fine. My machine seems to work pretty well with it, its just a little slow. I'm hoping to get a little more speed during cutting.

[Brady Watson]

John,
I recommend a 'real' power supply. You can piece together a torid transformer, bridge rectifier and capacitor(s) very inexpensively. You'll have to do your homework. Start here. (http://www.cnczone.com) If what I read in the link you posted is accurate, then you could use a 35v supply to boost speeds over that of the 24v supply you are currently working with.

My recommendation is to add the right supply for the job. Something that isn't going to let you down, cause a fire or overheat because it isn't properly suited to the task. PC power supplies are for PCs. They were not designed for running stepper motors. Yes...they CAN work...but are they really suited for a robotic tool? At one time this may have been the case...but given the myriad of options out there in the marketplace, I'd lean towards getting something a bit more robust.

-Brady

PS- I bought 2 torid transformers, 2 large can capacitors & bridge rectifier to power my Bridgeport CNC. This is a 72v/40A supply and it only cost me $150 to piece it together from NEW parts. Your setup would be considerably less.

[danhamm]

I am not sure just what you are doing with your
"toroid" transformers...but the only ones I am familiar with are used for switching in 12/24/48
volt dc...to...higher voltages..and wonder why
you would use them when ac is available..a normal
transformer is cheaper and more capable for this purpose..??? just my thoughts..and curiosity...cheers Dan

[Brady Watson]

Dan,
All your answers regarding Toroidal power supplies can be found here (http://www.plitron.com/Pages/advantag.htm)

-Brady

[danhamm]

Looks like I'm 20 years behind...thanks Brady...

[gerald_d]

John, here (http://www.campbelldesigns.com/files/power-supply-part-1.pdf) is a reliable guide for a CNC power supply. Ignore the first 3 paragraphs because your control board apparently limits the voltage to 24V. Notice that your transformer AC output voltage must be 1.4 times less than the desired DC output. The rectifier will increase the "apparent" voltage by 1.4 times. That link is to a page of this site (http://www.campbelldesigns.com/index.php).

[jseiler]

OK, I'm convinced. I think I'm going to send the switcher back and build my own linear. I found a 160VA 18VAC +18VAC toroidal from partsexpress.com for $38 and 25A bridge rectifier for $2.50. I need to find about 20K uF. Any good sources for one or two on the cheap? I guess I can use its output to drive a 24V fan for cooling (if necessary?). I take it this should draw about 1.3 A @120V and it looks like the power dissapation should be rather low, dumping a handful of watts in the xformer and a few in the bridge. I guess I could go with a higher VA rating in case I decide to use bigger steppers at some point, but I probably won't.

Thanks guys...its been a long time since I've been outside the world of 5V.

If I did the math right, if I wire the secondaries in parallel, I should get sqrt(2)*18 or about 25.5V @ 8A. Close enough. In AC world, 18*8 or 144VA. Since I've got 2A steppers, I should only need a peak of 6A, usually 30% less according to campbell's guide.

Thanks again,
John Seiler

[gerald_d]

I wouldn't go cheap on the capacitors - that size makes a big mess when they pop. Caps also go off with time - don't buy old surplus stuff.

[richards]

The debate between linear and switching power supplies is a little baffling. I've used both in my designs for over twenty-five years and greatly prefer the switching power supplies. The limitation that I've found with switching power supplies is that some of them require that you pull a minimum current all of the time in order for them to accurately output their rated voltage. Some of them have multiple voltage outputs with one output being the master and all others relative to it. In other words, if a supply has both 5vdc and 24vdc outputs, the 5vdc output might require a minimum current for accurate regulation and the 24vdc output would use the 5vdc output as a basis for its regulation.

On the other hand, a non-regulated linear power supply is going to be much larger in size than a switcher and it's going to have much lower efficiency. A regulated power supply will be even larger with the necessity of including heat sinks and fans to keep the output transitors (or equivalent) cool. I'm looking at one that I built about ten years ago to drive 24V steppers at 2 amps. That one used a voltage regulator, large transistors, large capacitors, and huge heat sinks.

My rule of thumb is to use a switcher whose output current rating is at least 50% higher than the current normally drawn by the circuit. Quality switching power supplies from reliable manufacturers have worked great for me.

[gerald_d]

Mike, my choice of power supply is the one with minimum components, lowest price, best availability of spare parts, least susceptible to dust and the one that I can troubleshoot myself when it goes wrong. Guess which one that is.


Some history (http://www.talkshopbot.com/forum/messages/26/1961.html). The cause of the failure seemed to be the small internal cooling passages that got clogged with sawdust. It was not repairable.

[dirk]

Gerald, I noticed you had posted that you were using a 80 Volt power supply on your new mods to your PRT. Will the stock prt motors handle this much voltage? I couldn't find specs on the stock motors, but I thought they were 2 volt. that would mean 50 volts max

[Brady Watson]

Dirk,
I have seen stock PRT motors running on 170v...the performance was very impressive to say the least. The 25x max voltage rule is a general guideline, and you should really look at a stepper's specs to see what it's demegnetization current/voltage is before applying this type of power to it.

-Brady

[gerald_d]

Dirk, I have these (http://www.orientalmotor.co.jp/cgi-bin/WebObjects/UPOMStep.woa/wa/F3?typeNameId=1&modelName=PK299-01AA&seriesId=2PK&frameSize=85) PK299 motors directly driving our early PRT. Later the PRT changed to PK296 (http://www.orientalmotor.co.jp/cgi-bin/WebObjects/UPOMStep.woa/wa/F3?typeNameId=1&modelName=PK296-01AA&seriesId=2PK&frameSize=85) motors through 3.6:1 gearboxes. The power supply voltage could have been higher than 80V but the (Gecko) driver electrics can only handle 80V max.

You can see on those spec sheets that the Voltage equals the max Current times the Resistance. The thing to watch is the max current - at high frequencies you can only approach the max current if you push the voltage way up.

[richards]

The theory behind using high voltage to run a stepper is to get the motor at its rated voltage as quickly as possible. I often use a 'bi-level' voltage supply to turn on rotarty solenoids. First, I send about 100VDC to the solenoid for just a few milliseconds and then hit it with a much lower holding voltage. Looking at the solenoid with an oscilloscope, shows a quick rise in voltage that never goes much higher than the lower voltage. The higher voltage helps to quickly turn on the rotary solenoid and the lower voltage allows it to stay on as long as necessary without overheating. The windings in a stepper motor are much like the windings in a rotary solenoid. Each device tends to resist any change in current. The 'chopper' drives for stepper motors, that use multi-voltage levels, handle the timing so that the stepper is not over heated.

Gerald, I normally agree with you on almost everything, but this time I would have to say that plugging in a switching power supply is about as easy as it gets. Adding to that the fact that most switchers available nearby cost significantly less than the components required to make an adequate linear supply. Both work. The switcher normally has an efficiency ratings of at least 75% while the linear power supply might run closer to 30%. When we're talking multiple steppers, that means a lot of power is being converted to heat with linear power supplies. Of course, if I did't have a switcher supplier within a few minutes of the shop and several switchers on hand, I might feel differently about it. (The Alpha stepper drives are really a combined switching power supply, chopper and sequencer that converts a simple pulse train into the voltages required for each winding of the stepper.)

[dirk]

I've purchased some used late model prt motors. their model number is A6497-9412KTG. I think this is a special model made for shopbot. I haven't been able to find spec sheet on this motor. It has a 4.39 ohm resistance so I don't think it is an equivelant to a PK296 as it's ohm resistance is 2.2. Anyone know the voltage of this motor?

[dingwall]

Gerald, did you switch over to Geckos?

[gerald_d]

In process, but havn't made sawdust yet. Link (http://www.talkshopbot.com/forum/cgi-bin/discus/show.cgi?tpc=312&post=26856#POST26856) to another post where I mentioned it.

[dingwall]

Hey - I took part in that thread. I must be scanning the posts too fast


I look forward to reading your progress.

[richards]

Gerald,
It looks like I wrote about power supplies by drawing too much on my experience and not enough based on the actual discussion. After looking at your posts about the Gecko drives and seeing just how simple the solution can be, I've re-thought my position. (In fact I just ordered a G212 drive to test on some old PH299 motors that have been sitting on the shelf for more than 10 years.) A transformer, bridge diode, capacitor and fuse is about as simple and reliable as it gets.

[ron brown]

Transformers, Bridge rectifiers and a filter cap have worked for me for many years. I also have a pretty high-tech switching power-supply, 35A @ 48V I power my ShopBot with and a hand-made torroid and filter cap as a third power supply. As far as I can tell the steppers don't seem to care.

Ron

[mikejohn]

With my PRT96 plugged into 230v single phase, how many watts is it pulling at full load, which I guess to be both x motors jogging at 2IPS.

I don't think its a lot, but I need to convince my new workshop landlord it's only small.
Alng the same lines, if there is an indicated watts value on hand tools, is this the maximum they pull?

..............Mike

[benchmark]

Hi Mike

The power consumption of the PRT96 is small compared to the router/spindle, vacuum hold down and the extractor.....have a look at the power supply in the PRT96 control box.

Paul

[dirk]

The PRT has a 300VA power supply. At 220V it pulls 1.36 amps. Thats probably less than the computer running it.

[gerald_d]

Yep, the hand tools draw no more than the indicated watt value.