Sinker EDM Machine
Hello,
I am in the early design stages of a sinker EDM. Using a PROP http://en.wikipedia.org/wiki/Electrical_discharge_machining
The axis is set up on linear rods, and motion is provided via a DC Motor. The Basic Principle of this machine is that the dc motor moves an electrode up and down on a ball screw. The purpose of the electrode is to use electricity to erode very minuscule amounts of material and flush them away with a dielectric fluid. This results in a very precise part with a surface finish resembling glass.
I would like to ask some of the members for there help and insight in any of the questions to follow.
Specifications:
________________________________
Voltage: 30 - 50 volts
Current: 15-40 amps
Frequency: TBD 1-10khz ?
24volt DC motor: w/ Encoder (haven't specified this yet)
Propellers Job (Not in any order of significance)
Monitor current
Monitor voltage
Set Frequency (arc pulse)
Set Feedrate (PID Loop)
Move DC motor (H-Bridge)
Monitor Encoder
Monitor Limit Switches
Function:
____________________________
As mentioned before, the electrode erodes a metal part. The electrode gap to surface has to be precisely controlled, if the electrode is touching would result in a direct short, too far and no arc builds.This is where the monitoring of the current and voltage becomes critical.
My intentions are this so far:
I would like to use the Quadrature Encoder and PID Controller Driver from the obex. This will read the encoder and set the feed rate.
I am a little unsure of how exactly to interface these high voltages and currents and the best method of going about it. From my research, it would appear that a voltage divider would be appropriate to bring the voltage down propeller tolerant. May be worth converting this to digital for use?
Current: Here is where i am really lost. I don't know how to safely read this large current. I have seen chips from Maxim that look like they would work, but i don't know enough about them to interface. Another thought i had, would be to put a very low resistance resistor inline with current flow, and then reading voltage on both sides of the resistor. This would give me an accurate current draw??? NO?
DC current will be provided from a toroidal transformer with the appropriate rating, and large caps to smooth the waveform.
PWM:
I have considered two methods of doing this.
First a large transistor of sorts would switch the dc power on and off rapidly giving me the rise fall I need at the work piece surface.
Second method would be to simply switch the main DC voltage to and from a capacitor, and simultaneous switch the charged capacitor inline with the work.
Because of the high dc current and voltage i seem to be having a trouble locating a transistor that would handle this type of environment and power dissipation in the event of a direct short. Is there some way to prevent this? Power Resistors?
I am interested in hearing any and all information that may help in my quest to a successful project. Obviously, safety is a high priority and i am willing to listen and learn. Nothing is to basic for me.
I am in the early design stages of a sinker EDM. Using a PROP http://en.wikipedia.org/wiki/Electrical_discharge_machining
The axis is set up on linear rods, and motion is provided via a DC Motor. The Basic Principle of this machine is that the dc motor moves an electrode up and down on a ball screw. The purpose of the electrode is to use electricity to erode very minuscule amounts of material and flush them away with a dielectric fluid. This results in a very precise part with a surface finish resembling glass.
I would like to ask some of the members for there help and insight in any of the questions to follow.
Specifications:
________________________________
Voltage: 30 - 50 volts
Current: 15-40 amps
Frequency: TBD 1-10khz ?
24volt DC motor: w/ Encoder (haven't specified this yet)
Propellers Job (Not in any order of significance)
Monitor current
Monitor voltage
Set Frequency (arc pulse)
Set Feedrate (PID Loop)
Move DC motor (H-Bridge)
Monitor Encoder
Monitor Limit Switches
Function:
____________________________
As mentioned before, the electrode erodes a metal part. The electrode gap to surface has to be precisely controlled, if the electrode is touching would result in a direct short, too far and no arc builds.This is where the monitoring of the current and voltage becomes critical.
My intentions are this so far:
I would like to use the Quadrature Encoder and PID Controller Driver from the obex. This will read the encoder and set the feed rate.
I am a little unsure of how exactly to interface these high voltages and currents and the best method of going about it. From my research, it would appear that a voltage divider would be appropriate to bring the voltage down propeller tolerant. May be worth converting this to digital for use?
Current: Here is where i am really lost. I don't know how to safely read this large current. I have seen chips from Maxim that look like they would work, but i don't know enough about them to interface. Another thought i had, would be to put a very low resistance resistor inline with current flow, and then reading voltage on both sides of the resistor. This would give me an accurate current draw??? NO?
DC current will be provided from a toroidal transformer with the appropriate rating, and large caps to smooth the waveform.
PWM:
I have considered two methods of doing this.
First a large transistor of sorts would switch the dc power on and off rapidly giving me the rise fall I need at the work piece surface.
Second method would be to simply switch the main DC voltage to and from a capacitor, and simultaneous switch the charged capacitor inline with the work.
Because of the high dc current and voltage i seem to be having a trouble locating a transistor that would handle this type of environment and power dissipation in the event of a direct short. Is there some way to prevent this? Power Resistors?
I am interested in hearing any and all information that may help in my quest to a successful project. Obviously, safety is a high priority and i am willing to listen and learn. Nothing is to basic for me.
Comments
Project Update
I bought a small tig welder to use as a power supply. It is capable of supporting 80amps at 30v. This should be more than enough current. The advantage to using a welder is it has built in short circuit protection and the necessary hardware to limit current. This will be advantages when it comes time to use a small electrode.
I have a found a few IC that are going to help me in my journey
http://www.st.com/stonline/products/literature/ds/9559/sty60nk30z.pdf for controlling the pulse generation
http://www.allegromicro.com/en/Products/Part_Numbers/0756/0756.pdf Current Sensor
http://usdigital.com/products/encoders/incremental/rotary/kit/e4p Encoder
I have included a picture of the machine in 3d. The black rod with the shape cut into is the actual electrode. The machine stands about 2 feet high and 10" wide. The blue translucent tank is where the dielectric fluid is caught and stored. The sprockets on top are faked in and will depend on the CPR I get on the encoder. I would like a resolution to about .00005". Seems like a very accurate number, but easily achievable with a gear reduction on the dc motor. The electrode does not have to move very fast. I would be happy if it could rapid its full travel in about 5 min.... Pretty slow.
I have been working on the PPDB and doing some little test circuits. I am mostly waiting on the chips to arrive so i can begin testing.