G203V and G100, recommended PS??

[nate]

Hello everyone!

I'm building my 3rd cnc machine now.

The first was a linley jig borer retrofit using 270oz steppers and Xylotex.
The second a router for a friend using 270oz steppers and Xylotex also.

For this next machine I am building om scratch a large 30" x 30" x 8" travel Mill and Plasma machine. It will primarily be a large Mill, with a bolt on plasma cutter setup and a drop in water table.

For this machine, I have decided to try Geckos. I like the 203V's for all the protection features. I also like the G100 due the the fact that it is a clean breakout board and Since I use Laptops mostly, I will run Mach 3 via an ethernet cable.

Now to the point of my post. Can anyone recomend the correct Power supplys or the G203V's and the G100? The mill will be a Gantry style machine, thus it will have 2 Y axis motors (slaved in Mach3)

The motors are Pacific Scientific 1.8 deg steppers. Powerpac nema 34 Hybrid 8 wire. 845 oz/in.

Any assistance with power supply selection would be appriciated. Also, recomended vendor would be helpful also.

p.s. where can I get the best deal on the 4 203V's and the G100? Does anyone out there offer a package deal?

Thanks,
Nate.

[nate]

Forgot to add-

Motors say-
Model K31SSFS-LEK-SS-02
5.5 A Bipolar series
35V
119W
1500RPM

[nate]

Anyone? would love imput here...please

[guru_florida]

I made my own power supply using a 35v toroid transformer, a large power rectifier bridge and some capacitors. It works great and powers my G203Vs without a prob.

This ebay item is close to what my toroid transformer looks like but not exactly: Search "toroid transformer"
http://cgi.ebay.com/60W-15V-15V-Toro...3286.m20.l1116

I assembled on a panel, not a pcb. I used simple crimp plugs to connect everything up electrically and mounted the transformer, caps and rectifier bridge directly onto the panel. I also mounted some terminal strips for my AC IN and DC OUT.

I used a 6800uF 80v cap on the output (my DC is 35v so 80v is safe). I distribute power to the G204Vs in a star wiring config (do not daisy chain power). I also placed a 2000uF cap right accross the G203v input terminal strip of each gecko. I did this because as each gecko demands power it will first get it from it's cap like a reserve tank, other gecko's should not see a drop in power. Many people post problems that are related to the power supply and not the geckos.

[Teyber12]

Originally Posted by guru_florida I made my own power supply using a 35v toroid transformer, a large power rectifier bridge and some capacitors. It works great and powers my G203Vs without a prob. This ebay item is close to what my toroid transformer looks like but not exactly: Search "toroid transformer" http://cgi.ebay.com/60W-15V-15V-Toro...3286.m20.l1116 I assembled on a panel, not a pcb. I used simple crimp plugs to connect everything up electrically and mounted the transformer, caps and rectifier bridge directly onto the panel. I also mounted some terminal strips for my AC IN and DC OUT. I used a 6800uF 80v cap on the output (my DC is 35v so 80v is safe). I distribute power to the G204Vs in a star wiring config (do not daisy chain power). I also placed a 2000uF cap right accross the G203v input terminal strip of each gecko. I did this because as each gecko demands power it will first get it from it's cap like a reserve tank, other gecko's should not see a drop in power. Many people post problems that are related to the power supply and not the geckos.

sorry for thread jack-

do you say don't daisy chain the power for the custom psu or in general? my keling psu only has 2 + outputs and 2 - outputs yet i have 3 motors- so i kind of have to

sorry for OT

[guru_florida]

Originally Posted by Teyber12 sorry for thread jack- do you say don't daisy chain the power for the custom psu or in general? my keling psu only has 2 + outputs and 2 - outputs yet i have 3 motors- so i kind of have to sorry for OT

In general. When I say daisy chain, I mean go from one motor to the next to the next. What should be done is to wire each motor back to the PSU and tie together there. I imagine that is what you are doing.The reason is that the wire from motor to PSU will have resistance. Even though this resistance is small it still *slows* the rate at which current can travel along it to the motor. If you daisy chain, you will have considerably larger lags in power from the motor's point of view. Instead, wire each motor back to the PSU and each motor will have an equal opportunity at the trough so to speak.If there is a single power line going to a 3x motor driver board then this is fine. The important point is that each motor has an equivelent "electrical distance" from the PSU, not that the wires must be seperate. (Should each motor wire also be the same length then? Well, ok, we are getting too picky here. However, High speed digital busses like in our computers *must* do this so each bit arrives at it's destination at the same time accounting for the length of the trace and the speed of electricity! If you look on a motherboard you will see lots of squiqqly line traces.) Long distance wiring (I'm talking in the range of 3' and more) become transmissions lines and thus you must think of each wire as being a small resistor and consequently the effect it will have on your entire circuit and it's power distribution.Good capacitor placement is recommended too. If you can, put your power capacitors as close to the motor driver power input as possible. No, not on the motor lines! Caps will supply an instantaneous power demand faster than the PSU, this is why we use them. The closer the cap is to the point of demand, the quicker the power will get there, thus less power brown outs will occur. This scales from supplying motor power, down to supplying ICs on a circuit board. In this case, only the scale is larger, but equivelently, we are placing large "decoupling caps" at key points of current demand.If you use Geckos or similar single amp drivers, place a smaller cap on each of the drivers instead of a large cap on the power supply. You get more benefit from smaller distributed caps than 1 large one. Still place a smaller cap on the PS output too. Basically, compute the total capacitance required, and divide by the number of motors +1. Put one capacitor each of this value on a motor and one on the PS. If you have a 3x or 4x motor driver, then divide by 2 and place a cap on the PS end and the motor driver power input end. And just to be clear, I am only talking about putting caps on power supply lines, not any signal lines or motor wiring lines.

[Teyber12]

wow great reply, thanks!

[Adamj12b]

nate:

I am going to try to help you out here, but it is a little tough without some information. After doing some research about your motor, I found some information. When you size the power supply for a stepper motor, you need to take the inductance of 1 phase and take the square root of it, then multiply it by 32. you did not have this information, so I had to go looking for it. None of the motors listed in the spec sheets list this motor directly, because it is a custom made motor. So I made a table of all the values for bipolar motors in the K31**** motor class. (See Picture 1) The best that I could figure, your motors have an inductance of either 8.3mH or 2.8mH. I came to this by using Ohm's law. you said the motors are 35V and 119W which yields 3.4A, but...5.5A and 35V yields 192W. Which is also confusing. So based of the formula of 32*Sqrt(mH): Sqrt(8.3)*32 = 92.2V and Sqrt(2.8)*32 = 53.5V. So and average of the two is 72.5V. See what I mean by its difficult without the inductance of the motor?

So moving on. The power supply needs to be able to handle all of the motors. For a power-supply you use the formula (80,000*I)/V where I = Current and V = Volts DC Wanted, to figure the optimal filter capacitor size. So figuring with the higher current value of 5.5A per motor just to be safe. 5.5*3 = 17.5, well round this to 18A. So (80,000*18)/72.5 = 19,862uF, rounded to 20,000uF. This would be the optimal filter capacitor value. To figure your output voltage to the motors, after the filter capacitor you would use the formula: Rectified Voltage * 1.414. So, In order to have a supply with an output voltage of 72.5VDC, you would need a transformer with a rated output of: 72.5/1.414 = 51.3VAC. This would then pass into a rectifier converting it to DC and then on to the capacitor for filtering and storage.

For recommendations: Kelinginc.com sells a power supply (KL-7220) rated at 72V and 20A. Well sized for running the motors at 5.5 A each and 72V. They also offer a 38V 9.6A regulated power-supply (KL-320-36). if you want to run them lower. Or even a 38V 15A unregulated supply (KL-3815) Due to the voltage drop on the unregulated power supplies, a few volts over is still acceptable. With a regulated power-supply, you can usually tune the supply back to meet the exact voltage you need.

As for the G100. I cannot recommend anything at this time. You could probably get a wall wart style power supply capable of providing 12V @ 1.5A at Radioshack.

I was looking at using the G100 at one point a few months ago for a knee mill conversion, but after talking Mariss, He recommended that I go with a smooth stepper instead. The G100 has lost alot of support under mach, and the smooth stepper has alot. The features that the SS support are rising high above the G100. I was able to get the Smooth Stepper with C23 Breakout board from CNC4PC.com for under $300. Thats 100 bucks less then the G100.

I hope this information helps.

guru_florida:

You said a few posts back that your toroid is 35V. This would produce 49.5V on the output, with the capacitor hooked up. If you meant to have 35V to your motors, I take it they are getting Hot? LOL. Im just curious. Also, What is the current of the motors you are running? By my calculations, it seems that even the additional 6000uF (2000uf X 3) caps, would not be enough to supply the large motors that 203V's can run. Were you having problems with them not getting enough power? I would love to see a picture of the cabinet you mounted the electrical stuff in. Sounds like it came out good.


-Adam

P.S. At full load (7 Amps) of a single 203V, The voltage drop of a AWG #16 wire,(Minimum recommended size by gecko) is .0861 Volts for a 3 foot long piece. For AWG #12 (The largest that will fit the the terminals) the drop is .0343 Volts for a 3 foot long piece. For #16, you would need a piece 38 feet long with a 7 amp load to get a 1 Volt drop. For #12, you would need a piece 89 feet long running at 7 amps to get just over a 1 Volt drop.

[nate]

Thanks everyone for the help.

Adam, thanks. I will study your post till I unerstand it better. I hve purchased the KL-7220 PS.

According to the Gecko Pdf for the 203v, i should be using a "j. 5.5A – 172.33K" current set resistor. this sound right?

[Adamj12b]

Yes, 172K Ohms is correct.

-Adam

[guru_florida]

Originally Posted by Adamj12b guru_florida: You said a few posts back that your toroid is 35V. This would produce 49.5V on the output, with the capacitor hooked up. If you meant to have 35V to your motors, I take it they are getting Hot? LOL. Im just curious. Also, What is the current of the motors you are running? By my calculations, it seems that even the additional 6000uF (2000uf X 3) caps, would not be enough to supply the large motors that 203V's can run. Were you having problems with them not getting enough power? I would love to see a picture of the cabinet you mounted the electrical stuff in. Sounds like it came out good.

Ah! You got me! I wasnt thinking...my DC is 32.5 to be exact. My Transformer is 25v, with the 1.4v voltage drop on the bridge rectifier it works out to about 33v. The DC voltage is right on for my steppers as per Geckodrive's FAQ.

I have 12'800uF on the PS now. I have 3 3A motors. I ordered some more caps but haven't placed them on yet. I thought I had the capacitance right, but recalculating it appears 22'150 is the correct capacitance. This seems rather high based on my intuition, that's a S**TLOAD of capacitance! What do you think? So far the machine has been running great! I did a part today and it was chewing through the aluminum smooth and with the most metal removal so far!

I'll get a PIC of the cabinet tomorrow.

C

[Adamj12b]

Hey guru_florida

22k uF is a large amount, but the math works out. The best way to check would be to use an oscilloscope on the main power cap to see the ripple while the machine is running.

-Adam