[freak_brain]

Can anyone tell me the proper way to fuse my electronics? I am using gecko G201's with 4.2V--4.7A steppers, 4 Wire. Do I fuse between the capacitor and the gecko on the "-" side as shown here


http://www.cnczone.com/forums/attach...achmentid=7925


or do I fuse between the geckos and the motors on each phase? or maybe both???

Also, I would assume if one is fusing the motor phases then you would use slow blow fuses slightly higher then the motors rated amps? For example, my motors (again 4.7A) I would use 5A slow blow on each phase, Would that be correct?


Thanks for any advice you could offer
-Allen

[gar]

050713-1515 EST USA

freak_brain:

Rather than answer your questions I want you think them through with the help of the following questions.

What is the peak inrush current to a stepping motor coil?

What determines the peak stepping motor coil current?

What is the shape of the motor coil current?

What is the thermal time constant of a motor coil?

What shunt current paths exist, or could exist?

At 150% of rating how long does it take a slow blow fuse to open?

What is gained by directly fusing a motor coil?

How or what is the relationship between the Gecko common and the control common?

What was the source of information that the creator of the referenced circuit used? Go to this source and determine if it is a reliable source for defining the location the indicated fuse?

Why would you want to fuse the motor leads directly if a fuse preceeds the Gecko drive?

See Littlefuse or Buss for fuse current vs time curves.

What is the thermal time constant of a typical semiconductor device compared to a motor coil?

Why do you use a slow blow fuse on a standard induction motor?
or At the input to your power supply?

Any time you are using a fuse or circuit breaker ask the question --- what are you trying to protect?

Making you ask and answer questions yourself is a technique that one of my old professors used. Most of our tests were open book and these were far harder to do than closed book. This prof would not answer any questions during a test, you were on your own. Even in non-test conditions he would try to make you analyze the problem yourself. He was the author of our textbook on electron tubes and created questions like what electric and magnetic fields were required to make an electron entering a hole on one side of a box spiral and exit a hole on the other side. The holes were not necessarily aligned.

.

[rboeser]

This may help some if you need a visual, compliments of Phils' website:

http://pminmo.com/simpleps.htm

[gar]

050713-1750 EST USA

There is a serious error in the circuit diagram of the previous post. Can you see it?

.

[rboeser]

The bridge has only the ac connection. The DC connection is missing/ connected directly to the AC line. Good thing it is fused...

[pminmo]

:tired: Good catch, man did I over look that one! :drowning:

Phil

[pminmo]

..........what mistake ?................

[GARG69]

the one thing they hammered into my head was a fuse is used to protect upstream components only.
hope this helps
gary

[gar]

050713-2010 EST USA

GARG69:

What do you mean by upstream?

If you mean the source, then I disagree. For example the circuit breakers in your house, shop, B1 airplane, or whatever are there to protect the wiring on the load side, or if at the equipment to protect the equipment.

A 20 amp breaker on a nominal #12 wire circuit (meaning reasonable length and not confined in a hot area) is selected to generally prevent overheating of the wiring and a possible fire. In my home with 200 amp service and a pole transformer of 100 kva and if there were no circuit protection, then shorting the end of the #12 circuit would not have much effect on the transformer or service entrance wiring, but would start a fire from overheating the #12 circuit.

By the way if you have aluminum wiring in you home get rid of it. This stuff is very dangerous and should never have been allowed by UL. Even if you are told you can coat connections to prevent high resistance joints do not trust this as a solution. You and your home are too valuable.

.

[freak_brain]

I guess I will need to ponder your questions for a bit, lol...


Thanks all
-Allen

[CJL5585]

Gar:
There is a serious error in the circuit diagram of the previous post. Can you see it?
----------------
1. 120 VAC line is not protected via a circuit breaker or fuse.
2. The drive power switch is in the power supply common side. (WRONG)
3. The fuse is on the power supply common side. (WRONG)
4. The lamps are rated @24 VDC -- Actual voltage to lamps is 33.9 VDC
5. The large capacitor should have a bleed off resistor across the + and - terminals to discharge the capacitor when the power is cut off. Otherwise, the circuit will be energized with the power off. Not really a shock hazard, but could cause you to blow up the Gecko drive if connecting or disconnecting motor, etc.

Note: Never put switches OR fuses in the negative lead. This is contrary to code.

The fuse for the Gecko drive goes in the + Voltage lead between the capacitor and terminal 2 on the Gecko. The fuse should be a 5 Amp fuse rated at 250 Volts (fastblow electronic fuse).

EDITED LATER after next post: On the other schematic, (Pminmo) the negative side of the supply is tied to chassis ground. Additionally, the transformer primary center tap is connected to chassis ground.

Jerry

[gar]

050714-2012 EST USA

CJL5585:

Jerry you did your homework on the first circuit presented by freak_brain. You have a very sharp eye.

Jerry my serious comment was directed at the pminmo post but no problem that you applied the comment to the first post by Allen. It brings up good talking points.

My question to Allen on what Gecko considered to be common for the chopper part of the circuit and whether this shared a common with the logic was directed at your point about switching the negative side. This really depends upon how the Gecko circuit is designed. If it is a don't care, then I would prefer to switch the positive side. This may also bring up the question of which, the negative or positive, side of the motor power should be tied to machine chassis and earth ground.

On the primary side as well as adding a fuse or circuit breaker I would insert an appropriately sized negative temperature coefficient resistor to limit inrush current and allow a lower rating on the protective device. Under steady state conditions it consumes some power, but it is a very useful tool.

Incandescent lamps of 24 v rating won't last long here. Also incandescent lamps on dc have a much shorter life than the same lamp on ac with the same rms current. The 757 pilot lamp was originally rated at 50,000 hours, but later reduced to 5000 hours because of dc applications.

If these are LED lamps then more needs to known. We use a high intensity Stanley LED for pilot light purposes using 20k ohms and 12v input. This same combination could run at 40v and not exceed the rating of either the LED or 1/4 watt resistor, at 40v the dissipation is 1600/20,000 = 0.08 watt. Even a higher voltage if you wanted to. At 40v the LED current is 0.002 amps ( 2 milliamps ).

It sure needs a bleeder.

.

[CJL5585]

Schematic in Post #1:
My concern was that Allen was using the schematic to wire his system and I was concerned because there were so many things that were wrong in the schematic.

When wiring Gecko Drives:

Let the power supply FLOAT. DO NOT connect either side of the DC power supply to chassis ground. Connect only the Gecko enclosure to chassis and allow the Gecko drive enclosure to act as an electrostatic or Faraday shield. The chassis is however connected to the earth ground connection from the utility company. I have built several drive systems using Gecko drives. If you connect the (-) common side of the power to chassis, one could end up with some hard to find ground loops, especially if one is NOT proficient in electronics and control wiring. It is recommended that the drive signals step, direction, etc. be run in separate shielded cables with the shield being connected to chassis ground on one end ONLY. Ground loops can occur due to switching power supplies, stray inductance, RF signals, etc.

Did not intend to tell you guys how to wire the system, but I do know what works for the life of the system without problems.

Jerry

[OCNC]

050714-2012 EST USA

On the primary side as well as adding a fuse or circuit breaker I would insert an appropriately sized negative temperature coefficient resistor to limit inrush current and allow a lower rating on the protective device. Under steady state conditions it consumes some power, but it is a very useful tool.

Here is the link to the data sheet for the NTC thermistor. The example given therein refers to a switch mode power supply. Could you comment on how one might use this for a linear supply? I'm assuming that it still comes after the bridge but since the voltage has already been lowered by the transformer how does one arrive at the correct rating?

Thanks.

Chris

[gar]

050715-0733 EST USA

CJL5585:

Jerry your comments are good and very usefull.

OCNC:

Chris:

The GE data sheet example could show the NTC either before or after the bridge with no effective difference.

When you have a circuit protector in series with a transformer input, then you want the NTC in this input side to limit the peak inrush from both the magnetics of the transformer and the filter capacitor. Do you want a second NTC that is on the transformer secondary side, I doubt it, but I have not made any effort to analyze this. Where ever you locate the NTC you are adding power loss.

We normally use a CL70 with a 175 va transformer. In an experiment with a CL70 it measured 16 ohms at 70 deg F. With no load on the secondary the CL70 read 1.6 v at 0.2 amps which is 8 ohms and the power dissipation was 0.32 watts. Putting a 68.6 watt resistive load on the transformer produced 0.667 amps and 2.0 v and 1.33 watts for the CL70. The maximum input current for the 175 va transformer is about 1.5 amps. We did not use a CL90 in our power supply because the output voltage rise time was too slow and caused microprocessor power on reset problems.

In picking a NTC you need to consider the NTC maximum current rating, and balance this with the amount of peak current limit, power dissipation in the NTC, and output voltage rise time.

In the switching suppy example the bridge is directly across the ac line and the goal is to protect the ac source, diodes, and capacitors from large inrush currents.

In a linear supply or some switching supplies you wll probably use a step down transformer. Here you have a magnetizing inrush current as well as the capacitor inrush current.

If you put an NTC just before a capacitor, then initially size this on maximum steady state current, backoff off to a lower reisitance of the same disc diameter to get desired output voltage rise time.

.

[OCNC]

Thanks GAR!

Chris

[gar]

050716-0858 EST USA

freak_brain:

Allen ---

Have you used Google to get data sheets on fuses and circuit breakers yet?

Try www.bussmann.com and look for data sheets on AGC, MDL, and GBB. There were no data sheets on GBB.

Circuit breakers --- Tyco (Potter & Brumfield) look at W6/W9 series. For example W67X2Q1-2-5. This is a nice on-off switch as well as being a circuit breaker.

Try drawing overlay curves of these for a 5 ampere rating.

What do learn from this information?

What can you tell us about the relative thermal time constant of typical semiconductors, and wire wound magnetic devices?

.

[gar]

050716-0915 EST USA

A suggestion on a bleeder.

I would eliminate the switch in the dc circuit unless there is a real need. I do not believe a need has been specified so far. Instead I would use the AC switch,

If we assume the capacitor is 30,000 mfd, and we want a discharge time constant of 1 second, then from T=RC we get R = 33 ohms. At 33 v this is 1 ampere and 33 watts. We do not want to continuously dissipate 33 watts and have this big resistor either.

Consider a normally closed relay contact in series with this bleeder and a 120 vac relay coil in parallel with the transformer primary. When power is off the bleeder is across the capacitor, when power is on it is disconnected. A 5 watt wire wound resistor should be more than adequate.

With a 5 watt resistor you could go down to 10 ohms and probably still be workable. This would make the time constant about 1/3 second.

You could also use an SCR as the bleeder switch,

Note: one time constant equals a 63% change in capacitor voltage.

.

[freak_brain]

WOW!! lots of info while I was gone.

First of all, thanks to all of you for your wonderful posts. You guys are way way way beyond me in the knowledge of electronics. I am very glad you're here because I am a machinist trying to make a machine so I have a lot to learn. Good thing for cnczone.

Okay, after reading all of this 12 times I came up with a couple of questions.

1. Putting a resistor across the terminals on the capacitor. I have heard of doing this to discharge a capacitor but I had not idea that I could do that and leave it on there. So how does one determine what resistor to use? I am lucky to understand the coding on the outside of the resistors let alone understand what the value of a resistor means.

2. I thought DC electricity flows from neg. to pos.? I assumed because of this that putting the fuse on the "-" side before the gecko would be correct. That is wrong?

3. "CJL5585" touched on a subject I had not even thought about, the sheilding on my motor phases. Each phase for each motor is wire and sheilded seperatly. One end of the sheilding should be connected to the chassis that my electonics are housed in?

I want this to be wired correctly so I am not at all offended with any advice or criticism you may have.

Gar..Thanks for the links to the data sheets. But I am sorry my friend, you are losing me...lol


Sincerely guys, I appreciate all of your assistance. It is a great help to guys like me who have little-no knowledge about electronics.

Please, just don't call me wife and let her know that I don't know everything, ok? :nono:


-Allen

[Al_The_Man]

means.

2. I thought DC electricity flows from neg. to pos.? I assumed because of this that putting the fuse on the "-" side before the gecko would be correct. That is wrong?

Actually electron flow IS from negative to positive, but an error was made, supposidely by the guy that flew the kite & keys in an electrical storm, he got it wrong and the world has now called it Conventional flow as opposed to actual electron flow.
It does not matter which side the fuse is in, only when one side is grounded as in a DC common or an AC neutral, then the 'High' side is fused. The electrical code states that circuits that are not referenced to ground should be fused in both power leads.
(you may have noticed that the diode symbol aka check-valve symbol appears backwards due to the Flow convention).
Al.