[PaulRowntree]

Hey Nick, it is good to have you back on the 'zone! First cuts look good!
Cheers!

[nickswimsfast]

Things to do this weekend:

- Rewire my homemade powersupply from 24VDC to 50 VDC.
- Resolder motors to a DB9 connector so I can use the gecko G540 I just received.

_____________________________________________
Where I'm at:

I just completed rewiring my powersupply to 50 VDC. I am troubled because now I am getting 50.7 VDC according to my multimeter. I need to drop the voltage somehow before wiring it to the Gecko G540. I'm thinking a lightbulb, or LED, or resistor would be a good way. It also would double as a powersupply "power light" so i know it is powered on.

Voltage drop
Electrical engineering is not my strong suit, so please correct me if I'm mistaken. I believe I can use this 25 Watt incandescent bulb to drop my voltage appropriately?

25-Watt Incandescent Green Light Bulb-144212 at The Home Depot

Here is my math:
P = IV where P = power (watts), I = current (amps), and V = voltage (volts)
25 watt bulb = (7 amp current) * Voltage drop

Voltage drop = 3.57 Volts


Thoughts? Is my math right? Will this put me in the 47-48 VDC range after the bulb?

Powersupply facts
Here are some quick facts about the unregulated powersupply:
- 50.7VDC @ 7 Amps
- Toroid Transformer specs: Antek - AN-5436
- 22000 µF50v Capacitor
- Bridge rectifier
- 7 Amp resettable breaker

[PaulRowntree]

Hi Nick;
I don't think the math is quite right here. A 25 W bulb has a resistance of ~500 ohms in the filament (P=IV, V=IR). The G540 draws 7A max, and so with a 48 V supply it looks like a ~7 ohm load. So most of the voltage drop will be across the light bulb, not the G540, and the power available in the G540 will be next to nothing.

Even if the numbers worked out with a different load resistor at 7A, it is dangerous to use a fixed resistance in series with a variable load (the G540) to drop the voltage across the variable load. WHen the G540 isn't working hard, it draws less current, so the voltage drop across the loading resistor goes down and the voltage across the G540 goes up. So at low current draws, the G540 could be over-voltage, and the magic smoke could be released.

Cheers!

[nickswimsfast]

Hi Nick;
I don't think the math is quite right here. A 25 W bulb has a resistance of ~500 ohms in the filament (P=IV, V=IR). The G540 draws 7A max, and so with a 48 V supply it looks like a ~7 ohm load. So most of the voltage drop will be across the light bulb, not the G540, and the power available in the G540 will be next to nothing.

Even if the numbers worked out with a different load resistor at 7A, it is dangerous to use a fixed resistance in series with a variable load (the G540) to drop the voltage across the variable load. WHen the G540 isn't working hard, it draws less current, so the voltage drop across the loading resistor goes down and the voltage across the G540 goes up. So at low current draws, the G540 could be over-voltage, and the magic smoke could be released.

Cheers!

Thanks for the feedback Paul!

I checked in with a buddy of mine who is better at this sort of thing, he recommended I use a bridge rectifier diode in order to drop the voltage. I found the following informative for these purposes:

Diodes
Diode - Full Wave Bridge Rectifier - YouTube

Bridge rectifier diodes are better rated for the conditions (i presume), and pull typically 0.7V across a diode. The rectifier I picked up from Radioshack hopefully will do the trick, and is rated beyond the voltage and current I am looking for. I hope that one diode will be enough, or I'll need to have a much better understanding of diodes to proceed.

I'll try putting one in series before wiring in the g540 and read the DC voltage.

On another note, I seriously dislike discharging the capacitor at 50V, it is much more dangerous than 24V... I either need to look into a bleed discharge resistor, or get a bunch of lightbulbs to discharge the system...


After going through all this dangerous hastle makes me wish I had just bought a powersupply from the get go, instead of building one. At least I've learned how powersupplies work

[PaulRowntree]

On another note, I seriously dislike discharging the capacitor at 50V, it is much more dangerous than 24V... I either need to look into a bleed discharge resistor, or get a bunch of lightbulbs to discharge the system...

Why do you need to discharge the caps? You never put a switch on the DC supply line to the G540, so the caps just do a controlled discharge through the steppers, no fuss or attention required. Mine takes just 10-20 seconds for the lights to go dead.

[nickswimsfast]

Why do you need to discharge the caps? You never put a switch on the DC supply line to the G540, so the caps just do a controlled discharge through the steppers, no fuss or attention required. Mine takes just 10-20 seconds for the lights to go dead.

Paul, that is very helpful!

I've only been working on the PS by itself, haven't started rewiring the motors to DB9 yet. (reconfiguring the transformer for 50VDC instead of the previous 24VDC)

Essentially, it's not hooked up yet. (Didn't want to run 51VDC into the g540 per the spec max of 50VDC.)

[PaulRowntree]

My old eyes and twitchy fingers had a tough time stuffing the stepper wires into the DB9 pin sockets. I have always disliked those things! One trick is to find some copper PCB material with the strip traces on them and multiple holes per trace. The gap between traces is exactly the 0.1" that is between pins of the DB9, so a 4-trace wide piece of the PCB gives a convenient soldering bridge between the cables and the pins. Watch out for shorts on the other side (where the resistor goes).

The DB9 shell is a bit of a bother though. Someone sells custom pcb bits for this purpose too.

[harryn]

It is good to see you posting on this build again. It is always an enjoyable read.

Perhaps an easier way to drop the voltage is to unwind one wrap of the secondary from the transformer. If it is fully potted in, then just take the wire on the secondary and counter wind it one wrap.

I am not a EE, but I have heard of others doing this, and it makes sense.

I am also playing with a wood frame design (for the same reasons you did). Do you happen to remember what engineering values you used for wood in your calculations ? Would you use the same value estimates now that you have had a chance to build it ?

In my case, I am attempting a design using plywood sheet, nominally 3/4 - 1.5 in thick.

[nickswimsfast]

Harryn, I'm glad to hear it, i'll make sure to keep it interesting

Thanks for the transformer idea, that would indeed be the best way. I'll take a hard look at it to see if I can do that.

To answer your question about analysis, the last picture in this post shows the values i used for my wood calculations:
http://www.cnczone.com/forums/cnc_wo...tml#post926211

I believe it was a fairly useful assessment, the only real way to validate it is to actually load your machine and measure the deflection then calculate the stiffness (or that of a sample part.) Which I may do later, and calibrate the analysis. You may want to run it with the same parameters to compare it to my machine, this will allow you to see how much stiffer your machine is

[harryn]

Thank you for pointing out those values. I had seen that post, but didn't enlarge all of the pictures for viewing. Next time I will know better.

Every design + build has it's strengths and limitations. I am actually amazed that you are patient enough to have redone the bolts so many times on those rails to make it all aligned. I am not that patient so another approach will be needed.

It has to be challenging to see commerical air frame construction methods and then come home to a wood frame project.

[nickswimsfast]

It has to be challenging to see commerical air frame construction methods and then come home to a wood frame project.

Yes, I'm impressed that you surmised that, I have been getting that feeling @ work. When I started this project I was in school I was a niave engineering student and had very little money to dedicate to the project, so it started out as a wood frame build out of necessity. The original project criteria is changing with the $$ part of the equation.

Motivation
I merely have come to terms with the idea that if I don't try, I won't know. So while a wood frame isn't ideal, it is low cost, easy to work with, and allows me to hit the ground running. I knew making a wood frame would make it a realistic goal to complete this project. I see the machine as a worthwhile stepping stone, as we all know on CNCzone... we love more projects!:cheers:

And a few of my favorite quotes to keep your gears greased:
“I have not failed. I've just found 10,000 ways that won't work.”
― Thomas A. Edison

“Genius is 1 percent inspiration and 99 percent perspiration.”
― Thomas A. Edison

Aligning rails on wood
Harryn, you likely already know this, and I'm sure you do... it was useful to establish one rail as a datum rail (bolt it down tight), and lightly attach the carriage plate to all the carriage blocks, and use the one rail to "self-right" the second rail. This helps you get close. As you predicted... I became good friends with my ratchet during the process for the fine tuning.

[nickswimsfast]

Significant progress:

1) Used additional bridge rectifier, and used two of the high rated diodes to drop the voltage. Due to supply variability, or reconfiguration of the wires, the voltage increased... Resultant voltage raised voltage just under a volt. (It should have dropped roughly 1.4 VDC)

Discharging that 22000 µF Cap @ 51Vdc enough times just scared the $$#@$ out of me (seriously like a gunshot.)

Video of discharging the Cap:
Capacitor Discharge - YouTube



2) Build new control box - After getting tired of working with the home made powersupply, I decided it was time to build a safer, and cleaner control box. This time, I bought a pre-built powersupply KL-350-48 48V/7.3A. It was a bit of a gamble, because there is pretty much no documentation or pictures of the device. This was perfect because the G540 is only rated to 7.3A anyways, obviously keling designed this with the g540 in mind. This PS has several nice features; led indicator, adjustable voltage adjustable pot, trickle discharge after device is powered off, and thermal operated internal fan. I wish i had dropped $50 on this from the start!
Switching Power Supply

I started looking around home depot, and picked up a Junction control box for about $33. A few hours later with all the components, and some perseverance with a dremel, I had all the components in. I was slightly concerned about melting the PVC plastic, and read online that its "ok" to machine, not sure if I believe it entirely. I opened the garage door just to keep the air moving. I bought an electrical connector kit to help aid with the connections. (very handy)

See the pictures for the build. I am quite pleased with it so far, and it powered on first go. Albeit, I was being much more meticulous about the setup, and sizing the wiring just right. I'll likely run the unit case open until I figure out how to add a fan in.



3) Motor DB9 soldering is next!


____________
Useful Electrical references:
http://en.wikipedia.org/wiki/NEMA_connector#NEMA_5
http://en.wikipedia.org/wiki/Electrical_wiring

[jsheerin]

Solder a large value resistor across the power supply capacitor. This will slowly bleed off the stored energy once you turn the power off. You can size it to change how long it takes to reduce the voltage and for how much power it needs to handle.
Bleeder resistor - Wikipedia, the free encyclopedia

[nickswimsfast]

Jsheerin, thanks for the info. I didn't know that is what they were called.

Fortunately, the new power supply I bought from keling has a built in bleeder resistor in it. After about 10 seconds from power down, the system is discharged.

Significant progress:

1) Used additional bridge rectifier, and used two of the high rated diodes to drop the voltage. Due to supply variability, or reconfiguration of the wires, the voltage increased... Resultant voltage raised voltage just under a volt. (It should have dropped roughly 1.4 VDC)

Discharging that 22000 µF Cap @ 51Vdc enough times just scared the $$#@$ out of me (seriously like a gunshot.)

Video of discharging the Cap:
Capacitor Discharge - YouTube



2) Build new control box - After getting tired of working with the home made powersupply, I decided it was time to build a safer, and cleaner control box. This time, I bought a pre-built powersupply KL-350-48 48V/7.3A. It was a bit of a gamble, because there is pretty much no documentation or pictures of the device. This was perfect because the G540 is only rated to 7.3A anyways, obviously keling designed this with the g540 in mind. This PS has several nice features; led indicator, adjustable voltage adjustable pot, trickle discharge after device is powered off, and thermal operated internal fan. I wish i had dropped $50 on this from the start!
Switching Power Supply

I started looking around home depot, and picked up a Junction control box for about $33. A few hours later with all the components, and some perseverance with a dremel, I had all the components in. I was slightly concerned about melting the PVC plastic, and read online that its "ok" to machine, not sure if I believe it entirely. I opened the garage door just to keep the air moving. I bought an electrical connector kit to help aid with the connections. (very handy)

See the pictures for the build. I am quite pleased with it so far, and it powered on first go. Albeit, I was being much more meticulous about the setup, and sizing the wiring just right. I'll likely run the unit case open until I figure out how to add a fan in.



3) Motor DB9 soldering is next!


____________
Useful Electrical references:
NEMA connector - Wikipedia, the free encyclopedia
Electrical wiring - Wikipedia, the free encyclopedia