Let's Make Robots!

Smurfs, Births & Negative Earths

Have any of you yet considered the idea of a robot’s elementary electrical arrangement being like that of a car?

What I mean by this is a negative earth system. As my robot’s chassis is entirely built from steel it would largely appear to be the most discernible idea. Perhaps though, I’m merely attempting to find territory more recognisable to my comprehension?

I’m sure we are all familiar with the concept of the electron being negatively charged and as such is drawn towards the positively charged terminal. So boasting a suitably solid chassis as a terminal to connect the ground of a component to will ensure excellent current flow.

I’m aware of potential for short circuits. However such could be negated by appropriate insulation. I’m also led to believe, the negative earth system itself greatly reduces radio interference. Which was a factor regarding its standardisation to automobiles, as this coincided with the rise in popularity of television.

Do any of you have thoughts on this?

Maybe I’m missing something obvious?

Any counters to this will be terribly appreciated.

I'd also like apologise for the blatant omission of both Smurfs and births which were promised in the title.

Thankyou,

Paul

 

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

bdk6 has it fairly well covered.

All I thought to add was that when I built my Schrödinger robot, his chassis was basically all metal, but I did not opt to use it for any electrical connections. My reasoning (besides the posible poor connections between metal pieces) was more related to the multitude of connections onboard and I considered it could increase the chances of a short, whereas if the chassis were isolated, then two wires touching the chassis at once would be needed to have anything short out.

You can indeed use the chassis as an "earth" or "ground" connection. Just don't forget later and hook something else to it that should not be grounded at that point. :-)

 

Oh, and here's one of those Smurfs you wanted.

 

Oh wait, could this be smurfs and births in the same cartoon?

                    NO, surely not Smurfette...!

                                 Ah, I see.

 

 

 

[ Picture interpretations by Dan M, with apologies to whoever owns the rights to the Smurfs. No money was received for these pictures.]

Mr. Dan M Sir,

Thankyou for the submission of a Smurf. It was wrong of me to miss-title a blog and for that I apologise to all who visited this article in the hope of reading about Smurfs.

Without your input Mr. Dan M Sir, my blog would still be Smurf free and titled less accurately. I will in future attempt more reasonable wording when naming my entries. Which should negate the requirements of yourself to intervene by providing me with Smurfs.

Thankyou once again Mr. Dan M sir,

Paul

In theory, the chassis ground is a fine (indeed very good) idea.  However, in practice there are issues.  As is true of almost everything in engineering, "it depends."  So here I will list some good and bad points, with varying amounts of commentary depending mostly on how much I feel like typing.

Good:

1.  Easy, convenient "grounding" point.  Yes, it is easy and convenient and might save a bit of wire.  

2.  RFI shielding.  Depending on the size, shape, configuration, orientation, and a few other things it may provide some RFI (radio frequency interference) shielding.  For this, it doesn't matter whether it is positive or negative "grounded" believe it or not.  I won't go into the why's of that.  Suffice it to say RF is a VERY tricky subject and RF engineers are often called wizards because of the black magic they wield.  It is also entirely possible that the chassis may actually increase received and transmitted RF energy.

3.  Safety:  There could conceivably be safety issues that a "grounded" metal chassis contributes to.  If done right, and there is some dangerous voltages/currents on your bot, it could help ( Your plug-in AC appliances with metal housings are almost always grounded.

NOTE:  In the first two points above and the first reference to ground in the third point, I quoted "ground".  In the last reference I did not.  That is because a battery operated device is usually NOT connected to ground.  Ground is the Earth.  The AC appliance is actually connected to Earth through a ground wire.  There is a huge difference.  RF is almost always connected to ground through capacitive coupling even when not physically connected.  RF is black magic.  Earth ground is often called exactly that.  A "ground" not connected to Earth is often called by various names, many times (confusingly) just ground.

Bad:

1.  Conductivity.  The metals used for a chassis are not as conductive as copper and will often (depending on many factors) have higher resistance than a copper wire.  The ground reference is the LAST place you want a current/resistance voltage drop.  The circuits we use for electronics usually operate on 5V or 3.3V, so a 0.25 Volt drop in the ground line can be disastrous.  

2.  Reliable connection:  The connections between different parts of a chassis, and the connections to components, are likely to be or become intermittent or high resistance.  Nuts and bolts and rivets or other mechanical fasteners aren't designed for electrical connection (usually).  Also, aluminum in particular oxidizes in free air EXTREMELY fast, and aluminum oxide is an excellent insulator.  It is also VERY hard (used for sandpaper and grindstones) so it is very difficult to breakthrough the oxide layer to get to the conductor underneath.

3.  RFI / unwanted signals ("noise"):  RF is black magic.  See above.  It is quite possible to get strange currents flowing through a metal chassis that cause all sorts of problems.  A metal RF shield does it's job by shorting the RF / electromagnetic current to ground THROUGH the conductive chassis / shield.  That current is free to go into your circuit that you are trying to shield, actually causing much more harm than good.  There are other even more esoteric factors.  RF is black magic.

It should be noted that, yes, cars DO use the chassis as one terminal of the power supply, usually negative.  That works very well with non-critical circuits like lights, starter motors, vent motors, etc.  But take a look at any modern car with lots of sensitive electronics and you will find most of those circuits have a separate ground wire running to them, often all terminated in one central location.  They will also have lots of protective circuitry on the board to protect the sensitive stuff from the nastiness that shows up on the car power supply, including the "grounded" chassis.  In short, I do NOT recommend using this as your "ground wire".  It is not a bad idea ot ground it to provide some shielding, then run the "ground" wire of the shield all the way back to the battery connection, not to any other place in the circuit.  But I would not use it as the ground connection for any sensitive electronics.

Mr. Bdk6 Sir,

Thankyou for the full and informative points. You have certainly aligned, what in my case at least may have been buried somewhere deep in grey matter.

The issues you speak of with respect to RF are certainly interesting and could explain the concerns I appear to be having with my television signal. I will seek the aid of an exorcist to rid my set of the black magic affliction which appears to have taken hold.

Before any decisions are made, I will further study the subject with particular reference to the areas you have raised.

I thankyou for taking the time to provide me with these reference points,

Paul

Ah, Mr. 6677 Sir,

Good to see you again. So you know not of any hazards regarding this system? Excellent, then I shall proceed with my design.

Thankyou,

Paul

I don't know any hazards no, but at the same time I am not an expert so you might want someone with a bit more authority than me in these matters to give an opinion. My only thought is like yours, shorting.

 

I have indeed heard about cars using it though. When you read guides on how to jump start a car for instance there appears to be much debate over whether to connect the negative cable on the car to be jumped to either a metal surface or direct to the battery terminal. I have always used the battery terminals and have never seen the sparking people mention as being hazardous. Only time I saw any sparks was when getting confused over my battery polarity (I am used to the idea of red pos and black neg, so seeing a brown and black off of the same battery led me to think brown = faded red = pos, it was neg, BIG SPARK)

Mr. 6677 Sir,

Thankyou for conceding to suggestion of further opinion. I always respect a man who doesn’t attempt to create a fallacy regarding his knowledge.

My reasoning for the desire to create my robot’s electrical system with such a design is purely for familiarity. A system I’m accustomed too appears to make more sense than learning a second.

 

Jumping a vehicle’s engine does raise a certain debate, however speaking as a gentleman who’s jumped more vehicles than most, let me assure you, more often than not connecting directly to the battery will produce improved results. An inferior contact can occur when using anything else, due to corrosion, engine lacquer, miss-decanted oil, etc. Although many new cars are manufactured with the battery concealed, so a direct connection isn’t always exactly undemanding. Granted there will often be motivation for another to dispute this.

The vehicle to be jumped should have its headlight switch in the on position before connecting the cables, as this will draw any initial current surge away from more delicate electronics. The headlights ought to be deactivated once the cables are connected. The donor vehicle must remain running throughout the whole process.

The wiring colour you speak of is not standardised across vehicle manufactures. Certain motor companies such as the French brands and Vauxhall take pleasure in corrupting a system all are familiar with. For this reason I would always advise on cross referencing the polarity with the symbols on the battery.

I hope this may be of some assistance to you in a future state of affairs.

Thankyou,

Paul

 

Its funny you should mention Vauxhall, was a Corsa. I've labelled the cables now, I like big obvious indicators. My first car (1999 Skoda Felicia) had a few battery issues. I became quite intimate with the jump cables or bump starting, bumping a car from a tow rope in traffic when you've had a license for 2 days is the way to live (or not to live, I forget). Sadly that car died in April.

Mr. 6677 Sir,

A great bond exists between a gentleman and his first automobile, comparable to the friendship between a child and his dog.

I remember my first well, a mark 3 Ford Escort 1.6 Ghia. She was finished in metallic quartz gold and first registered in 1984. She was on her second engine, third gearbox and fifth carburettor. The engine’s breather system would blow a mixture of oil and petrol into the air box which would in turn drain out onto the exhaust manifold and ignite in spectacular fashion.

She gave me my first taste of freedom. I named her Erika and she’d take me anywhere I wanted to go, occasionally she’d take me back as well.

These days when I drive a car, whatever it may be, however refined, or luxurious it may be, I think back to Erica and realise nothing can put a grin on my face like she could.

Paul

 

So your suggesting using the chassis as a ground? Seems a legitimate theory to me. Keeps a few ground wires out of the way at least. Don't know if I'll ever be trying it but certainly something to think about