Page 1 of 2

Wacky Idea: Tube Based HW Implementation

Posted: Wed Mar 21, 2012 4:22 am
by Fred
Anyone feeling bored? It'd be youtube worthy to run a modern COP/Sequential 4 cylinder from a fully tube based EMS with just the CPU and basic peripherals done in a more normal way :-) You could likely use some old school V reg too. Huge old 630V caps. No point at all, purely for amusement value.

Someone: Do it.

Fred.

Re: Wacky Idea: Tube Based HW Implementation

Posted: Wed Mar 21, 2012 10:22 am
by Fred
[06:36] <ToxicGumbo> http://www.tube-tester.com/sites/nixie/nixie-tubes.htm
[06:54] <ToxicGumbo> http://www.dailymotion.com/video/k3ZXmW0CS5bdo9ryFm

That old guy is amazing, I love fine craftsmanship like that.

Re: Wacky Idea: Tube Based HW Implementation

Posted: Wed Mar 21, 2012 12:44 pm
by micyek
I've been lurking around for about a year or so, but I just had to sign up to reply to this thread. I chuckled when I read your idea. All I could think of is the big box of tubes and things on the hood of the Delorean in Back to the Future III.

By the way I'm very intrigued by FreeEMS, it's a great project...even though I'll probably never get a chance run it. LOL

Re: Wacky Idea: Tube Based HW Implementation

Posted: Wed Mar 21, 2012 3:49 pm
by Fred
Yeah, there are a LOT of lurkers like you were until now. They should all stick up their hands and say Hi :-)

Great memory, I'd forgotten that detail of that movie :-)

I think we need a higher res picture such that anyone doing this knows what to aim for:

Image

BTW, I stalked you and noticed two things, well, three, two identical reasons why you might not get to use FreeEMS and that you don't have this: http://www.facebook.com/FreeEMS

Welcome along! :-)

Fred.

Re: Wacky Idea: Tube Based HW Implementation

Posted: Tue Jan 22, 2013 1:03 pm
by fausto
Non practical, but it would be a really cool show-off thing. Imagine it installed on a very old rat-rod, with all that old relays noise pulsing the injectors...

Re: Wacky Idea: Tube Based HW Implementation

Posted: Tue Jan 22, 2013 2:58 pm
by bradtpt
Fred wrote: [06:54] <ToxicGumbo> http://www.dailymotion.com/video/k3ZXmW0CS5bdo9ryFm

That old guy is amazing, I love fine craftsmanship like that.
It's a sad thing but the reality is that time will see it that skills such as will go the way of history.

A thorough craftsman, really enjoyed watching him work. It's refreshing to take a step back from our fast paced, tech ridden lives sometimes and reflect on where we've come from.
fausto wrote: Non practical, but it would be a really cool show-off thing. Imagine it installed on a very old rat-rod, with all that old relays noise pulsing the injectors...
Relays wouldn't be fast enough... BUT I believe it could probably be achieved on an engine with a very poor VE (hence we need minimal dynamic range in terms of fuel delivery) and the smallest injector that would run the application, so we had a low load injection time of say >6 ms. I have never measured the deadtime of a relay :) With some accurate injector characterisation data (static flow, deadtime vs voltage, we wouldn't need low pulse non linearity as we'd size the injectors so we never had to go under say 5ms) and a secondary relay deadtime table (lol!!) I believe the fuel delivery could be modelled and the output stage driven with relays.

Hang on, did I just start discussing HOW this could actually be done? Hmm, that's enough internet for me for one night I think... :?

Re: Wacky Idea: Tube Based HW Implementation

Posted: Wed Jan 23, 2013 8:09 am
by Fred
No need for relays, valves are more than capable of driving injectors directly or perhaps through an isolation transformer. This could definitely be done, it would be costly, though. The scale of it would mean it would remain compatible over various fw versions, so no need to wait if anyone wants to get in neck deep :-)

Ignition could be more of a challenge if ignitors weren't used. Thoughts on that? :-)

Never enough dreaming! :-p How do you think we got where we are? ;-)

Fred.

Re: Wacky Idea: Tube Based HW Implementation

Posted: Wed Jan 23, 2013 11:03 am
by TheRevva
2 * EL34 valves using a custom wound output transistor might be able to (just) do the job of the ignition.
(Assumes you don't need more than about 100mJ of spark energy)
Hell, since you'd need a custom wound output transformer anyway, you could make it COP! (Or rather, CNP since they'd be rather heavy and would get a little warm).
With CNP, and a V8, that's 16 * EL34s so we better specify a bigger alternator too just to keep the damn heaters warm! (Wire each pair of heaters in series directly across VBatt)

Not quite sure how you want to derive the anode voltage on them. Considering the technology, it might be considered RUDE to use a silicon SMPS?

Hmmm, perhaps I'm showing my age here...
I still fondly remember building a stereo amp with 8*EL34 (same as 6CA6 in USA IIRC) a few decades ago.
Each channel used 4 valves arranged as 2 pairs of push pull.
The output transformers were HUGE, HEAVY, and rather expensive

Re: Wacky Idea: Tube Based HW Implementation

Posted: Wed Jan 23, 2013 7:49 pm
by Nick_nl
I made a nixie clock once, it does look nice. But the end result was pricey.

Re: Wacky Idea: Tube Based HW Implementation

Posted: Fri Feb 01, 2013 2:10 am
by fausto
For ignition, I had an idea: Maybe something like flyback drivers of old TVs may work. I remember my grandpa had an ultra old DC tube TV, but I was unable to find any schematic of the flyback driver.