Diode in a Bullnose

This first project is a modern take on some fairly ancient technology.

Pictured below are Pat and Barry Goulding with Pat’s 1923 Morris Oxford – also known as a “Bullnose”.   The car is running fine mechanically but the electrical system has not been charging the battery for some time.

The switch-box which controls the electrics of the car is recessed into the dashboard.  It is shown with the engine running after the fix was completed, hence the charge registering on the ammeter.  The switch on the left selects engine run, stop, and battery charge; the light-switch on the right turns on either sidelights or headlights.

The wiring diagram below which we found somewhere describes the innards of the box.  It is laid out to represent the actual copper interconnections so you may go blind trying to figure out the circuitry.  It took me quite a while to realise what should have been clear all along – the node marked “cut-out frame” is connected to the frame of the device in the top-left called the cut-out.

What this cut-out is essentially about is that if you simply connected the dynamotor (combined starter and generator) across the battery, then current would flow between the two almost continuously – from the battery to the motor when stopped, thus cranking the drive; or from the motor to the battery when under-way which would be fine for charging but not for trying to stop.

What is required is a button for connecting the battery temporarily for cranking the engine, i.e. the starter switch, and a device which will only connect the dynamotor to the battery when it is ready to supply current.  How this cut-out works is that it takes current from a shunt winding on the dynamotor and directs it through the weak coil (left) which pulls the cut-out contact shut.  This completes the main charging circuit between the battery terminals and the dynamotor, but it also brings into circuit the strong coil of the cut-out (right).  This provides a hard-latching of the contact thus preventing it from opening until the current drops below a much lower value (in control theory this property is termed “hysteresis”).

Of course there is a very simple modern solution to this problem (definitely not one for the vintage purist though) – a 2-terminal device called a diode which allows electrical current to flow in one direction but not in the other.  A subtype of this is called a Schottky diode which we can use here; these consume less power and thus stay cooler, but I’ll leave the gorier details for the footnote.

The diode (actually two diodes in one package) is shown below with a heatsink attached to dissipate as much heat as possible.  Leads have been soldered on to the legs for attachment to the switch-box with the anodes in red and the cathode in black.  Also shown is a modern automotive fuse called a MAXI – it is almost the same size as the original in the switch-box but unfortunately slightly too high to fit under the cover.  Nevertheless, the filament can be cut out and screwed in to the original fuse holder for a more accurate and reliable version; a 20A was originally fitted, and these can be found in the MAXI form-factor.

The picture below shows the wiring of the diode to the terminals at the back of the box – the cut-out and its wires have already been removed.  The anode connection is to the centre terminal of the motor switch while the cathode connects to the ammeter terminal on the lightswitch side, both with standard crimps.  Long leads are left on the diode for routing out of the switch-box to somewhere with free air movement, as confining the device in a closed space could cause continuous heating during charge.

Finally, the unit was refitted back in to the dash.  The diode can be tied up behind the dash out of sight when it has been tested.  In this case the current ran quite high at first, above the 10A range of the ammeter, so some adjustment was made to the dynamotor to trim back the current supply.  Whether this adjustment mechanism was jammed or no longer functioned we could not get the output to go much below 10A at idle, so we found it necessary to keep the sidelights on to take some of the load.  The battery is a modern SLA car-battery so over-charging would shorten its life and in an extreme case cause the to rupture.  This is more of an issue with the dynamotor itself though so we’ll leave that for another day.

The gory details

I have included the calculations for selecting the diode and heatsink here for anyone who might be interested.  Firstly, the baseline requirements are:

  • Maximum reverse voltage (Vrrm): 15V
  • Maximum forward current (If): 20A

When these conditions have been satisfied with some margin, the next most important figure is the forward voltage (Vf) as this will dictate the power consumption and heating of the diode while charging- hence the use of a Schottky which is characterised by low forward voltage.  The part STMicroelectronics STPS60L30CW was chosen to easily meet these requirements:

  • Vrrm: 30V
  • If: 30A
  • Vf: 380mV

The next step is to calculate the power and heating of the diode while charging – the power is given simply by

P = V • I

so this works out at 7.6W for a maximum charging current of 20A.  The thermal resistance of the diode to its case, Rth(j-c), is given as 0.8°C/W, so for 7.6W the temperature difference between the diode junction and the case will be around 6°C.  If we assume pessimistically that the ambient temperature where the diode is sited is 70°C, then this leaves a maximum temperature difference of 74°C (maximum junction temperature of diode: 150°C – 70°C – 6°C) between case and ambient.  This implies a maximum thermal resistance of around 10°C/W.  The chosen heatsink is given here and provides a lower resistance of 7.2°C/W.

Note that the key values of current and ambient temperature were chosen very pessimistically, so we can neglect the resistance of the thermal paste between the diode and heatsink.  The temperature rise at the diode junction in the typical case should be quite modest – around 30°C.

4 Responses to “Diode in a Bullnose”

  1. Hi, Just checking if you have any photos showing my good side?

  2. Nice work; the cut-out is really an EM diode. But it’s operation is even more subtle that your (abbreviated) description.

    The “weak” winding, by virtue of its intrinsic DC resistance (sometimes with an external resistor with a smaller tempco than copper) is a voltage sensor. When the generator output voltage reaches that required to charge the battery, the contacts close. The reduction in air gap provides hysteresis. The generator output is now the same as the battery voltage, for as long as the contacts are closed.

    What if the generator output goes down, so that its open-circuit voltage (if the contacts were open) would be lower than required to charge the battery? That’s what the “strong” coil is for. Under those conditions, current flow (in generalized terms, you may have a positive ground) is backwards; the generator is taking power from the battery. The strong coil is a current sensor. When the current flow is backwards, the strong coil “bucks” the weak coil, opening the contacts.

  3. Thanks for the detailed explanation Ted. I knew something diode-like was going on but didn’t examine the operation to that degree!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: