[modeleng] Re: Topsy - to charge and how to charge, that is the question!

  • From: "R.L. Roebuck" <rlr20@xxxxxxxxxxxxx>
  • To: "'modeleng@xxxxxxxxxxxxx'" <modeleng@xxxxxxxxxxxxx>
  • Date: Wed, 30 Mar 2005 12:54:10 +0100 (BST)

On Wed, 30 Mar 2005, Peter Sheppard wrote:

> Hello Rich,
>
> Its a 4QD unit, their "Uni" controller.  The motor is shunt wound, but I
> grabbed the reversing contactors when I removed it from the stacker truck so
> reversing is not a big issue.

I would bear in mind issues like - damage to the controller when you
switch from one direction to the other:-

1) if the motor is running, due to the resulting voltage spikes
2) if the motor is stationary, due to the static charge that the insulated
motor windingscan carry ralative to the systems earth.

...I'm sure 4QD have advice leaflets on it, but I would consider fitting:-

1) a freewheels diode BEFORE the direction changing switch
2) some static dissipating 1MOhm resistances between the motor and
controller sides of the switch.

As far as the controlling the alternator situation is concerned - is
it safe to say you are considering two main options - when it is charging
the battery and when it is not charging the battery, but surely the
charging levels aren't going to be constant unless you use some kind of charging
regulator. For example, if you switch it to charge the battery, but the
battery is fully charged, then the engine will be very lightly loaded, and
may rev its way into oblivion; but without having the throttle set this
high, it may stall if the battery is almost completely discharged.

My gut feeling is that a slightly more sophisticated form of control may
be required, which I suspect is what Alan was getting at some weeks ago.

Some kind of governer, to adjust the throttle to keep the revs level when
charging the battery, but that when the battery voltage reaches a certain
level, the revs are backed off.

If the engine is as comparatively underpowered as you say, then it may
well need the revs backed off to allow for some cooling off time.

>
> I haven't looked closely at the alternator, I got it from a lorry scrapyard 
> and
> was only too pleased to find that it worked!
>
> If I get a chance I'll have a look, but I don't remember anything obvious.

Have a look, but if not, there are ways and means of determining ratings -
let me know what you find, and if you need a few pointers.


Yours,


Rich.
PS One thing above all, with this kind of setup - given the complexity and
somewhat uniqueness - bear in mind that:-

1) you maybe ought to set the whole thing up on a bench and check that
your idea works as you expect it will - possibly leaving out the 4QD
controller at first, as this is the most likely and expensive thing to get
damaged, particularly by voltage spikes and static electrcicity.

2) make sure when you build the loco that you allow for motors and
alternators getting damaged - ie if that nice 24V alternator lasts all of
5 minutes, and you got a replacement one - is it likely to fit in.

3) If you have an oscilloscope handy, check to see if the 28volts from the
alternator is smooth - if there are voltage spikes coming from it, then
these may well destroy your 4qd controller quicker than you can say
'please pass me my cheque book and a large theraputic hammer'.


>
> Cheers
>
> Peter
>
>
>
> On Wednesday, March 30, 2005 12:14 PM, R.L. Roebuck [SMTP:rlr20@xxxxxxxxxxxxx]
> wrote:
> > Hi there Peter,
> >
> > Can I ask - how are you acheiving motor control in all of this. Do you
> > have a PWM unit lined up such as 4QD supply, or are you thinking of using
> > a large resistance, or something else?
> >
> > Also, do you have any details from the makers/ratings plate on the
> > alternator?
> >
> >
> > Yours,
> >
> >
> > Rich.
> >
>
>
> MODEL ENGINEERING DISCUSSION LIST.
>
> To UNSUBSCRIBE from this list, send a blank email to,
> modeleng-request@xxxxxxxxxxxxx with the word "unsubscribe" in the subject 
> line.
>
>

MODEL ENGINEERING DISCUSSION LIST.

To UNSUBSCRIBE from this list, send a blank email to, 
modeleng-request@xxxxxxxxxxxxx with the word "unsubscribe" in the subject line.

Other related posts:

  1. Home
  2. modeleng
  3. Archive
  4. 03-2005