Solid State Power Distribution Modules - Clearly The Future!
Discussion
I think you’ve cut yourself a bigger slice of humble pie than you deserve Dave.
What you appear to be suggesting is that by using more modern technology, it should be possible to improve the reliability of our TVRs and in particular, the wiring - I think that’s a given.
You also conceded that the product we need does not as yet exist - so no harm done - nobody is going to read this post, rush out and buy the bits and do it this weekend. What’s more, if you’re starting out with a dodgy alternator, variable resistance earth connections and a distributor, perhaps there are some other projects you would be well advised to consider first…
Solid State Relays are not more unreliable than Electro-Magnetic Relays - but they do have different vulnerabilities which need to be taken into account. EMR contacts need protection when switching inductive loads to prevent them arcing or welding-up which in effect is the same end result as an SSR failing in a latched state. That doesn’t stop us using them, we’ve simply incorporated the protection into the circuit design. The same applies to SSRs. Spikes and noise are factors to be considered at the design stage but it’s not a deal breaker. Given that we’re only switching 12 Volts in relatively simple configurations, building in some immunity to spurious triggers is hardly rocket science.
In my line of work we use solid state switching almost exclusively - because quite simply and unquestionably, it is more reliable, for longer - and that could save lives. However, because each installation has a degree of customisation, we have to satisfy our clients that the quoted MTBF (mean time between failure) of each individual component can be expected to meet the contracted design life of the assembly. To ensure that the component delivers on it’s quoted MTBF, the design has to ensure that each part is always operating within it’s specified performance envelope. Which is a fancy way of saying that you don’t ask anything to perform outside of its comfort zone. Quite simply if SSRs are failing on a regular basis then they are either wrongly specified or are being subjected to harmful conditions. That’s human error at the design/specification stage and quite often happens when simple substitution takes place rather than a properly conceived design.
I think a modular design would also be the way to go for at least two reasons. Not only would it make repair a simpler and less expensive issue, it would also make it feasible to custom build and configure the best possible assembly for each application - and easily alter/upgrade it. I also think there is really interesting potential for some switching options we have not been able to consider before. The ‘current’ arrangement of fuses and EMRs is quite limited compared to what could be achieved with a solid state design with quite basic programmable logic. If each SSR switch can be configured to be either normally open or normally closed, latching or momentary and have dependency on any other relay or input, the permutations become almost endless.
The potential market for this is not confined to TVRs or even other classics. It could be deployed in all manner of small scale applications whether in competition cars, specials, boats etc. in fact anything where a reliable, scaleable, configurable control system is required with built-in and resettable protection.
So as you say Dave, lets hope that someone out there spots the commercial potential for this as an adjunct to their ECU business - and from a purely personal perspective, I also hope it’s Mr CANEMS.
I do think that this kind of thing will come in time and it will almost certainly (IMO) be CANBUS based for the widest possible application but vehicles are a bloody awful environment to develop for - where I work now we did the 'premium' HiFi setup for Bentley (a £5k tickbox when you spec up your new Bentley) and I know that even that was a nightmare for handling all the external noise and also making sure that what you make doesn't affect anything else ...
Before my current job I used to look after a lighting system that gets used in a lot of high end domestic and commercial installs (we did the lighting control for Wembley Stadium - 4,500 controlled circuits) which - again - is bus based with switch inputs, logic processing and various kinds of outputs. I wrote my own lighting and heating control system for my old house myself using that and the functionality within that could very easily be used to do a virtual vehicle using sensor and switch inputs - none of it is particularly complex and, to be honest, implementing control in logic is far easier than physically interlinking circuits so the technology is there and it's all very well established and used in all sorts of fields ...
... it's just making it rugged enough and reliable enough for the stupidly harsh automotive environment - I've often wondered about building my own microcontroller setup for the Tiv (I even got a basic vehicle electrics lights / indicators / horn / wipers etc. control setup working on breadboard) and I'd love to implement it as it was easy to do stuff like automatically putting on the headlights when the wipers were turned on or when it started getting dark (including allowing for street lights), doing offside / nearside parking lights when the indicator stalk is flipped when the ignition is off and so on - it's all great fun - but the reality is that I wouldn't trust it in an automotive environment.
Phil
Before my current job I used to look after a lighting system that gets used in a lot of high end domestic and commercial installs (we did the lighting control for Wembley Stadium - 4,500 controlled circuits) which - again - is bus based with switch inputs, logic processing and various kinds of outputs. I wrote my own lighting and heating control system for my old house myself using that and the functionality within that could very easily be used to do a virtual vehicle using sensor and switch inputs - none of it is particularly complex and, to be honest, implementing control in logic is far easier than physically interlinking circuits so the technology is there and it's all very well established and used in all sorts of fields ...
... it's just making it rugged enough and reliable enough for the stupidly harsh automotive environment - I've often wondered about building my own microcontroller setup for the Tiv (I even got a basic vehicle electrics lights / indicators / horn / wipers etc. control setup working on breadboard) and I'd love to implement it as it was easy to do stuff like automatically putting on the headlights when the wipers were turned on or when it started getting dark (including allowing for street lights), doing offside / nearside parking lights when the indicator stalk is flipped when the ignition is off and so on - it's all great fun - but the reality is that I wouldn't trust it in an automotive environment.
Phil
Slightly different idea because this guy replaced his wiring harness for weight saving and its a Jaguar XJS but is an interesting read, and its all about MPG so ChimpOnGas will love it 
http://www.theregister.co.uk/2014/10/19/crawling_f...
http://www.theregister.co.uk/2014/10/19/crawling_f...
AceOfHearts said:
Slightly different idea because this guy replaced his wiring harness for weight saving and its a Jaguar XJS but is an interesting read, and its all about MPG so ChimpOnGas will love it
http://www.theregister.co.uk/2014/10/19/crawling_f...
Crikey, he's even sadder than me http://www.theregister.co.uk/2014/10/19/crawling_f...
I simply went with one ECU controlled fan and the other one triggered by the otter switch, this gave me the staggered fan effect on the cheap.
No spike from both fans chiming in at the same time means no monetary idle drop, this has also allowed me to fractionally lower my idle speed.
And with the Canems closed loop idle feature seamlessly catching any other engine load based RPM drops (like the power steering being on full lock) means my idle is solid now as a rock, always.
No spike from both fans chiming in at the same time means no monetary idle drop, this has also allowed me to fractionally lower my idle speed.
And with the Canems closed loop idle feature seamlessly catching any other engine load based RPM drops (like the power steering being on full lock) means my idle is solid now as a rock, always.
Gassing Station | Chimaera | Top of Page | What's New | My Stuff