Home-brew overdrive logic controller

[JohnC]

We're still in lockdown here in Sydney, and idle hands the Devils work do (as my old religion teacher told me many times). So I decided to design my own overdrive logic controller. I've always liked the intent of the logic controllers out there, but I prefer to brew my own. I also like to try to do so using components which were available when my car was manufactured. So here's my effort. I really would appreciate any suggestions for improvements (@Lebro - looking at you!).

My intent was this:

• Overdrive only available when inhibitor switches allow it (obviously)
• Overdrive is selected/deselected  by a momentary switch.
• Overdrive is deselected whenever there's a gear change, and needs to be manually selected again afterwards if so desired.
• A warning light is illuminated when overdrive is selected.
• The overdrive warning light is dimmed when sidelights are on.

Here's the circuit I came up with. The CD4013 flip-flop IC was first available in 1968, and may well have cost as much as the TR6, but I think I've satisfied my intent to be "period". The circuit works in the LTSpice simulator, and on a breadboard in the car  (including the warning light) but I haven't done the work to smallify it onto a soldered board. My next step is to get out the soldering iron, but I thought I'd seek critical input first. Attached is the schematic and the Spice file in case anyone wants to play with it.

Cheers,
John

 

 

 

Edited September 27, 2021 by JohnC
Schematic removed - Updated circuit attached to later post

[RobH]

 

Question John - Q1 doesn't seem to do anything, what was your intention for it?  As long as there is power via Q3  it is always conducting.  It looks to me as though the LED D1 should work just the same if it was taken direct to earth and Q1 omitted. 

I confess I can't really make out what you have done with the interlock switches as they are not shown. I see the note saying the earth for U1 is removed when the interlock opens but does that mean the Vss pin remains connected to the overdrive relay ?   Have you provided a commutation diode for the overdrive relay coil?  If not, U1 might not like the back-emf spike generated when power is removed. 

 

[JohnC]

Hi Rob,

Q1 only allows a current (through D1) if its base is energised by an emitter current from Q3, which in turn is only turned on when there's a high signal from the Q pin of the IC. The overdrive relay is represented in the circuit diagram by R10, and is also energised by the transistor Q3. The Vss pin is simply not earthed when the o/d interlock switches are open, and so the whole circuit is inert. Does that clarify things? I confess I'm not sure what back-emf spike will be generated.

Cheers,
John

[RogerH]

Hi John,

I have a Logic design that has worked very well for ages. Except one that I made appeared susceptible to BEMF from the solenoid.

The circuit is identical to those the have never had a problem. It may be that this particular solenoid it a bit naughty.

The BEMF from the solenoid can be quite destructive.

 

Roger

PS - I have tried to use semiconductor but they tend to be rather iffy in the car electric system - very noisy.

Mine is based on relays and cocks a snoot at rough electric.

 

Edited September 26, 2021 by RogerH

[RobH]

57 minutes ago, JohnC said:

Q1 only allows a current (through D1) if its base is energised by an emitter current from Q3, which in turn is only turned on when there's a high signal from the Q pin of the IC.

Yes - so really Q1 isn't needed because the power to the lamp is switched the same way. As I see it, when Q3 is off there is no power to the lamp anyway. 

  The voltage spike generated by a  relay turning off can be quite big - up to hundreds of volts sometimes - and semiconductors can be damaged by it.  A diode wired in parallel with the coil, oriented so it is reverse-biassed and does not conduct with the applied current, will short-circuit the spike and protect circuitry. 

[SpitFireSIX]

Nice work JohnC. Nice to see another semiconductor approach instead of the relay approach.

[Steves_TR6]

Interesting!, not had time to review properly yet but love the idea !

steve 

[JohnC]

10 hours ago, RobH said:

Yes - so really Q1 isn't needed because the power to the lamp is switched the same way. As I see it, when Q3 is off there is no power to the lamp anyway.  

The voltage spike generated by a  relay turning off can be quite big - up to hundreds of volts sometimes - and semiconductors can be damaged by it.  A diode wired in parallel with the coil, oriented so it is reverse-biassed and does not conduct with the applied current, will short-circuit the spike and protect circuitry. 

Aha! The penny drops. Thanks Rob - you're absolutely correct. I designed the circuit in bits and changed my approach a couple of times. Q1 was  leftover from one of those changes. Didn't do any harm but completely unnecessary!

And I'll add a diode as you suggest. Updated schematic and Spice file attached.

Exactly the input I was looking for in posting this, even if I was a little slow on the uptake

John

 

 

 

OD logic device v2.pdf OD logic device v2.asc

Edited September 27, 2021 by JohnC

[JohnC]

10 hours ago, RogerH said:

PS - I have tried to use semiconductor but they tend to be rather iffy in the car electric system - very noisy.

Mine is based on relays and cocks a snoot at rough electric.

I couldn't work out how to get the toggle on/off working with a momentary switch using relays. Please would you post your circuit? BTW a friend of mine has pointed out I could achieve the same using a tiny Arduino compatible processor. That might be my next project, although it wasn't available when the car was built. I wonder if it would better withstand the car environment?

John

[RogerH]

Hi John,

PM sent

 

Roger

[RobH]

Roger's circuit for operation from a single push-switch is especially cunning, and uses a minimum of robust parts to achieve it.

 Just my opinion of course but it seems the modern way is to reach for an Arduino or similar PIC as first resort, perhaps because no-one is taught to work using discrete components these days.  It is a good solution for complicated control situations or where the function needs to be readily changeable. However, for a relatively trivial problem like this and where the function will be fixed forever, I believe a PIC is considerable over-kill and may need special protection measures to make it reliable in the 'hostile' electrical environment of an old car.