PDA

View Full Version : dc amps vs ac amps



Lee Roberts
02-09-2008, 01:59 AM
what is the differece between ac and dc amps if any?

Lee Roberts
02-09-2008, 02:00 AM
I was told:


DC Amps and AC amps are the exact same thing, they are the measurement of electrons past a given point, the difference is that the electrons of AC go back and forth (alternating) and DC go only in one direction (direct).

Measurement must however be done with two different meters, if you measure AC with a DC ammeter it will twitch a lot but you'll get no reading, if you measure DC with an AC ammeter it might just let it's magic blue smoke loose and you will have a nice christmas tree ornament but no more ammeter.

So after comparing a few other sources on the differences between, AC and DC i'm happy that the above is true.

The reson for my original question regarding amps was because i am currently selecting the correct wire to use as supply lines to each of my driver cards. To futher explain that, i was already aware that diffrent gauges of wire would also draw current and if that is the case i needed to also ask the original question above.

Hopefully this post will also help somone else in the future, who may be looking for the same answer!

What i would like to know now is, how do we determine when a wire will start to draw current?

The reson i'm asking this question is because when i had my out building wired by an electrishion, he told me that you can't use over rated cable because it would draw, build up, demand current itself. So if thats the case really to do the job right i need to make sure that the wire i use to supply each driver, isnt going to do the above.

Maybe you guys could help me with this, am i going "way overboard" for what my circuit will be doing? or is this a good show of consistency in doing the job right.

What i'm basically getting at is this, my PSU is currently supplying my circuit with 38VDC @ 14a. Each of my driver cards are rated @ 4.2a (RMS 3A) MAX, so we know that the driver cards combined could require 12.6a (4.2 x 3 = 12.6).

The wire i have decided to use is rated @ 15a, so 15 - 12.6 = 2.4. So my wire should handle what is being requested in ampage but will the wire pull a little more current from the PSU as its rated @ 15a and a little over what the peak will be, going back to what i said above about wire and the (draw, build up, demand) ?

Now moving onto fuse's to protect my driver cards, i understand it that fuse selection is done so by adding 25% to the peak rating. So 4.2 + 25% = 5.25, so i would need to use a 5.25a fuse to protect each driver but actually a 5a fuse is the closest i can find so thats what i will be using? and if its true that the wire will (draw, build up, demand) do i need to take this into consideration on my fuse selection.

Lee Roberts
02-09-2008, 04:16 AM
So reading back on what i'v posted/asked above, i obviously know that mains current is generally allot more then what we are playing with here but shouldnt we be applying the same procussions/standards regardless, if we are to protect ourselves and/or our equipment.

Any help and/or advise is much appreciated!

A Note to mtompson: I'm using the same drivers and PSU as you are, however i'm using two PSU's wired in parallel to give me the 14a.

Robin Hewitt
02-09-2008, 03:37 PM
You have two things to consider, resistance and capacitance. Fat wires have lower resistance so you lose less power heating the cable. Try to keep the wires between controller and stepper short otherwise the capacitance can affect the current sensing in a chopper circuit. Fat wires have more capacitance but not so as you will notice.

A 5 Amp fuse will handle more than 5A but may tire if you flow 5A all the time.

The great AC/DC debate was thrashed out in America by Edison and Tesla.

Edison's problem with DC was it meant a power station on every city block, but as he pointed out, you can't run a motor on AC.

Tesla responded by inventing an AC motor that we still use today.

Edison said the higher AC voltage was dangerous and arranged for criminals to be executed electrically to prove his point.

Tesla won but blew all his money trying to transmit power without wires.

Edison lost but left a corporation behind who told everyone that he was cleverer the Tesla.

Tesla left no corporation so he lost by default even though he was actually much smarter.

OTOH Tesla has a fundamental unit named after him but we measure nothing in Edisons :D

irving2008
03-09-2008, 08:34 PM
CheekieMonkies

Wire doesn't 'draw' current, it resists the flow of current. To use an oft quoted analogy, which isnt perfect but works, think of the wires as a pipe, the power supply as a pump and the 'amps' as the flow rate of the water. The voltage of the supply is what drives the system... a bigger pump ( = more water pressure) can push more flow (amps) through the system, but ultimately the size of the pipes, the wires, is what limits the flow... too much and the pipe bursts (overheats/melts). Which is exactly what a fuse does... to fix the point of overheating at a safe location.

Some thoughts.... if you have 3 motors at 5A do you need wire capable of carrying 15A? No, you need wire capable of carrying 7.5A because each motor should be seperately wired back to the power supply via its own 7.5A fuse, never daisy-chain high current motors. Fuses are rated at the current they will carry without blowing, which should typically be 25% higher than the highest current you expect to encounter. Generally wire should be rated 50% higher than the expected maximum so that the fuse becomes the weak point. It doesnt hurt to use thicker wire but it costs more.

Stranded wire is preferable to solid core wire for most applications where vibration is present as it is less likely to fracture. Also for high frequencies and for systems where there are fast transients (like a stepper driver) stranded wire is better as high frequencies travel in the outer skin only so more strands gives a better result (more surface area).

To wrap up - the difference between DC and AC current is - there is none, except that AC current reverses direction in a (generally) cyclical form, whereas DC flows in the same direction all the time. However, consider the power dissipated (energy lost) in a wire. If you have 10meters of cable it may have, say, a resistance of 0.1ohm so a 10A DC current would would dissipate 10W of energy (and incidentally drop 1 volt, so a 50v PSU would give 49volts at the other end). A 10A RMS AC current would do exactly the same, however the peak current would be 14.14A (10 x square root of 2). Why? Because the RMS (root mean square) only truely applies when the shape of the AC current follows a sine wave - or, as in the case of a microstepping drive, averages out to be the same as a sine wave. A higher peak value is required so that the energy transmitted per second is the same as the DC value. It follows therefore that if there are circuit components sensitive to the actual instantaneous current value, like a fast acting fuse or circuit breaker, then they will trip at a lower AC current than the seemingly equivalent DC value because of this higher peak.

Incidentally a typical low-cost AC volt or ampmeter is only accurate when the AC flow is sinusoidal, as it is calibrated to show the RMS value against the actual peak value measured.

Hope that all makes sense.

regards,
Irving...

Lee Roberts
04-09-2008, 02:19 AM
ok thanks guys, so Irving where i have used 15a wire here for all my supply wires for my fuse's, i need to upgrade the 1 main wire feeding each fuse to 50% of what my expected maximum could be.

So in that case x3 4.2a drivers could ask for 12.6a down that one "supply" line, so adding 50% i need to upgrade that supply line to a 18.9a wire?

Am i right on this? or do i need to expect my maximum to be what the PSU is giving out, so 14a i would need to use 21a wire, please see the pic to understand more what i mean.

Thanks,
CM

irving2008
04-09-2008, 12:17 PM
ok thanks guys, so Irving where i have used 15a wire here for all my supply wires for my fuse's, i need to upgrade the 1 main wire feeding each fuse to 50% of what my expected maximum could be.

So in that case x3 4.2a drivers could ask for 12.6a down that one "supply" line, so adding 50% i need to upgrade that supply line to a 18.9a wire?

Am i right on this? or do i need to expect my maximum to be what the PSU is giving out, so 14a i would need to use 21a wire, please see the pic to understand more what i mean.

Thanks,
CM

CM,

Do not daisy-chain the fuses back to the PSU, wire them individually back as per the diagram below....

The power supply won't be able to supply more current than its maximum rating. depending on the design of the supply it will either go into current limit or will blow a fuse :)

Lee Roberts
04-09-2008, 08:33 PM
Ok mate, my PSU says "Peak" 9a.

Will you explain why its not a good idea to daisy-chain the fuse's ?, ill also pop this thread in the FAQ section as i think its going to be a good one for others to read.

CM

Robin Hewitt
04-09-2008, 08:54 PM
Why do you want fuses anyway? I don't use them, just one more thing to go wrong, or in this case, 3 more things to go wrong :D

Lee Roberts
05-09-2008, 03:07 AM
Irving, i just found this bit of information can you confirm:


it is not a good idea to "daisy-chain" the supply wiring from one drive to the next, as current drawn by a drive that is nearer to the power supply can effectively "starve" drives of current lower down the chain. In fact, if the wiring from the power supply to the drive is likely to exceed about a foot (300mm), it is advisable to fit local reservoir capacitors of 470 microfarads across the supply terminals of each drive - these must have a voltage rating at least 25% higher than the supply voltage.

Robin Hewitt
05-09-2008, 09:47 AM
Perhaps "starve" is a bit excessive :D

A 1 Farad capacitor would deliver one Amp for one second, so .00047F is good for about half an Amp for 1 millisecond.

Your stepper driver should have built in decoupling to handle fast transients. I think maybe you are getting bogged down worrying about things that aren't going to give you problems.

You can add fuses and fat caps later if you find a need, but I don't think you will.

irving2008
05-09-2008, 11:23 AM
While what you are proposing CM will work, good electrical working practice is to independently feed high current circuits directly from the power supply to minimise interaction between circuits.

The reasons are many:
1/ A good quality power supply is a low impedance which will absorb transients. Daisy-chained circuits potentially allow transients from one driver to impact on another, for example the back-emf from the driver into the PSU when the motor coil is switched. This can push the supply voltage up several 10's of volts and it may be that only the PSU's impedance and back-emf protection is protecting the other drivers.

2/ Starvation is possible, if for example the connection between motor 2 and 3 isnt as good as that between motor 1 and 2 (the feed being to #1) and motors 1 or 2 are drawing a high current then the voltage across motor three may drop several volts leading to poor torque regulation. However we only really need to consider this possibility where the resistance of the connection is likely to be significant in respect of the current drawn - traction systems for instance where the current is measured in tens or hundreds of amps.

3/ Step current demands are transients and the shape of the step is determined by the higher frequency components which by their nature travel in the outer skin of the conductor. Long conductors present a much higher impedance to transients and therefore can benefit from localised capacitive reservoirs. Without the right equipment (osclliscope for instance) its impossible to tell whether these are actually needed. One determining factor will be the regulation capability of the power supply... i.e. how the output voltage changes under repetitive step current loads. A 5% regulation is considered poor, 0.1% would be pretty good (thats 2.5v or 0.05v on a 50v supply).

You say your power supplies are rated at 7A typical, 9A peak and you are running them in parallel. What supplies are they and are they linear or switched mode regulated or unregulated? If unregulated or switched mode you may find extra capacitance at the motor beneficial (but again it can be hard to tell).

Make sure the wire connecting the supplies together is heavier than the motor wire - best practice is to wire both supplies to a common connector block from where the motor wiring is taken (in high current systems a copper bar (busbar) is even better). Failing this, use 2 or 3 pieces in parallel.

I would argue that fuses are essential. If they 'go wrong' and are correctly sized then its a sign there is a problem somewhere... better to blow a fuse than fry a driver...


Incidentally, I just re-read your original post re your 'electrician'... I reckon he was talking out of where the sun don't shine!

Lee Roberts
05-09-2008, 03:48 PM
You can add fuses and fat caps later if you find a need, but I don't think you will.

Hey Robin, that quote i did in my last post was just somthing i found on a guys build log website, i wont be caping anything :beer:

Lee Roberts
05-09-2008, 09:25 PM
While what you are proposing CM will work, good electrical working practice is to independently feed high current circuits directly from the power supply to minimise interaction between circuits.

Ok thank you, i just wanted to now and understand/learn rather then just be told :).



You say your power supplies are rated at 7A typical, 9A peak and you are running them in parallel. What supplies are they and are they linear or switched mode regulated or unregulated?


They are from MCP, PS407:

The PS Power supplies are available as 40V and are non regulated switch mode power supply specificaly made for use with our range of stepper and servo drivers.

Supply Voltage: 180 to 250V AC


Output Voltage:

PS407 - 40V DC at 0A and 38V DC at 7A
Output Power

PS407 - 260W (Continuous)

Make sure the wire connecting the supplies together is heavier than the motor wire - best practice is to wire both supplies to a common connector block from where the motor wiring is taken (in high current systems a copper bar (busbar) is even better). Failing this, use 2 or 3 pieces in parallel.

I will look at how i would like to go about connecting the supplies together and get back to you with what i propose doing, could i use copper clad board as an option? reson being i'v got some of the green screw terminals you find on the BOB boards i could solder in.


I would argue that fuses are essential. If they 'go wrong' and are correctly sized then its a sign there is a problem somewhere... better to blow a fuse than fry a driver...

I couldnt agree more, while the drivers dont cost a fortune, it would be much more inconvenient replacing one of them, then a fuse located on a front panel deisgned with easy access in mind ;).

I'v just found this as well: http://www.mycncuk.com/pdf/wiring.pdf

CM

Robin Hewitt
05-09-2008, 09:33 PM
I would argue that fuses are essential. If they 'go wrong' and are correctly sized then its a sign there is a problem somewhere... better to blow a fuse than fry a driver...

If the fuse goes pop because the driver has fried, you are going to buy a new driver anyway. They are so cheap they aren't worth repairing.

If a fuse goes pop for no good reason, the part you are cutting becomes junk PDQ. Then the cutting tool ploughs into a holding clamp, the bed dances a jig and it can get really expensive really quickly.

You don't want to end up using both hands to stall it while trying to switch it off with your foot :D

Lee Roberts
05-09-2008, 09:49 PM
If the fuse goes pop because the driver has fried, you are going to buy a new driver anyway. They are so cheap they aren't worth repairing.

If a fuse goes pop for no good reason, the part you are cutting becomes junk PDQ. Then the cutting tool ploughs into a holding clamp, the bed dances a jig and it can get really expensive really quickly.

You don't want to end up using both hands to stall it while trying to switch it off with your foot :D

If the fuse gos "pop" then the driver stops as dose the motor as dose the axis in question ?, isnt the fuse there to protect the driver from becoming fried?

irving2008
05-09-2008, 11:13 PM
If the fuse gos "pop" then the driver stops as dose the motor as dose the axis in question ?, isnt the fuse there to protect the driver from becoming fried?I think what Robin is implying is that the fuse wont blow as fast as the driver could fry and that's certainly true for certain situations but if you short the motor wiring by dropping a screwdriver on it or the motor overheats and internally shorts a turn the fuse will still probably blow before the driver's MOSFET fails cos of thermal runaway unless it is very tightly spec'd...

irving2008
05-09-2008, 11:26 PM
Ok thank you, i just wanted to now and understand/learn rather then just be told :).

No problem, happy to help

They are from MCP, PS407:


I will look at how i would like to go about connecting the supplies together and get back to you with what i propose doing, could i use copper clad board as an option? reson being i'v got some of the green screw terminals you find on the BOB boards i could solder in.

I'd be careful of those green things, most of the 0.1"/2.5mm spacing found on BOBs are only 1 or 3A rated. You'd need the bigger 0.2"/5mm spaced versions which are 7 or 10A rated

What I've done in the past for some high power radio gear is taken a length of brass bar, 8mm x 4mm, drilled the ends 4mm clear to fit the PSU terminals (3.2 or 3.6mm screw lugs) and then drilled/tapped 3mm (or similar sized BA) take off points for ring terminals along the bar (use brass screws)

I couldnt agree more, while the drivers dont cost a fortune, it would be much more inconvenient replacing one of them, then a fuse located on a front panel deisgned with easy access in mind ;).

I'v just found this as well: http://www.mycncuk.com/pdf/wiring.pdf

CM
Yes I've seen that diagram before... note its a connection diagram not a wiring layout diagram - it shows what connects to what, not how to do it properly.