sofakingdom

iTrader: (1)

Seems like there's ALOT of people on these forums that have defective tachs...

Would there be interest in a complete all new PCB, all new electronic parts (assuming I can find some LM1819 chips... otherwise I'd have to have a functional "core" one to re-use) with GM's crappy design problems fixed?

You would have to send me your old defective PCB in any case, so I could retrieve the plug-in connector pieces off of it; but no other parts would be re-used. All new resistors of a CORRECT and ADEQUATE power dissipation rating so they wouldn't burn up, all new caps, new ICs if I can find them, new bare board of a completely different design but all the same connections, pre-calibrated, switchable among the various cyl #s. You'd just assemble it to your existing meter, which hardly ever goes bad.

No idea what the price point would be; let's just say for the sake of argument, $100. Who would ACTUALLY BUY one at that price? Keeping in mind the old saying, "money talks, BS walks"... I don't want to hear a bunch of P&Ming about "can't afford it", if that's all you're gonna do then just keep it to yerself... if ya wanna play, ya gotta pay. What about $50? $25? the more of you guys that commit to buy up front, the less I'd have to charge.

I have a 82-88 or so Camaro one as my example. No idea how different the ones for Firebirds are, or for the later model cars. If somebody can send me non-working boards for those I'll be able to know what I can do about them.

Again, no fixed point where the price would end up yet, just gauging interest.

Amillionoh7

iTrader: (1)

have a 92 tach i can send to ya.

sofakingdom

iTrader: (1)

OK cool, thanks.

Amillionoh7

iTrader: (1)

pulled tach today I will send it tomorrow. do you want the tach and pcb to make sure they work together?

sofakingdom

iTrader: (1)

Sure, if you have the whole thing, that'll allow me to test all the individual parts for compatible intechangeability.

Scorpner

iTrader: (4)

I think that starting from scratch is the way to go on something like this. I'm interested to see what you come up with.

sofakingdom

iTrader: (1)

The silence so far has been deafening. Amillionoh7, thanks.

As an update, I'm looking at doing it some other way besides the old discontinued chip. The principle that the meter works on allows many other approaches. Cost would be more or less the same; just wouldn't have to worry about finding stuff that's been out of production for 20-odd yrs already.

No point in continuing design effort unless there's actual interest, though.

ndndndnd

Well, my tach is pretty much dead and I have no idea how to revive it.

I suspect the only way I would proceed would be to either attempt to replace all the componentry individually, with all the issues that brings with my ignorance of electronics and scarcity of the required components, or to retro-engineer a new solution… and I have almost no idea how to do that.

So, I'm game. For me, buying a PCB from you would probably be the most economic and least stressful solution.

someone972

iTrader: (2)

Probably wouldn't be hard to design a PCB around CS8190 or other newer tach driver chips. Took a peek and only the 20-pin SOIC version of CS8190 is immediately stocked, but if you're doing the soldering yourself I'm sure you could source the 16-pin DIP package as well (might just be a wait?). My tach is working fine right now but I'm all for more solutions to common problems.

ndndndnd

I think that kinda proves my point! I have no idea what you're talking about…

However, a quick google did bring up this link:

http://www.dinoplex.org/tachoconversion/

which does look a bit like a 'how to' guide…

I guess with our cars, the tricky bit about making a new PCB would be retaining the oil pressure gauge?

ndndndnd

I've been reading around this a little bit - and I've just further convinced myself that I don't really know what I'm doing!

However, apparently there are PCB creator programmes where you can design your own PCB and then send it off to get made (Eagle PCB Creator?) therefore I guess making a replacement PCB (including oil pressure gauge) is relatively straightforward...

Has any got any experience of these programmes and would be able to design a replacement tach PCB to put up for everyone to use? I'd have a go, but my skills are limited and my car is at the exhaust shop for the next couple of weeks!

sofakingdom

iTrader: (1)

That's what I do for a living some of the time.

Having circuit boards made in small quantities is NOT CHEEEEEP. It's kinda like (not EXACTLY this, but...) you pay $1,000,000 in cash; and you tell em, do you want 100, 1,000, 10,000, 100,000 or 1,000,000 boards made. Then you sell em based on a cost TO YOU of $10,000, $1000, $100, or $10 $1 each. The ACTUAL PCB ITSELF doesn't matter, within certain bounds.... it's all about THE QUANTITY. It's why high-$$$ test equipment is made in the US, UK, Japan, or Germany; but toys, and Harbor Freight, are made in China, Vietnam, India, or Thailand.

So if our cars use, say, 4 different styles of PCBs (let's say, 82-8x Camaro, and 8x - 92 Camaro, and 82 - 8x Firebird, and 8x - 92 Firebird) that just MULTIPLIES the whole "volume" problem by a factor of 4. No, actually... it PENALIZES the Bird guys WORSE, because there's less of them to begin with, there being less Birds in the world than Camaros. Interest levels may be different (so far as I can tell, that's ZERO - NOT ONE SINGLE PERSON has committed to me that they'd buy a tach PCB for ANY kind of car at ANY price, even FREE, although one has generously provided help) but demographics, in the end, rule.

Yes EAGLE is one of the PCB design suites that I've used. It's a CHEEEEEEEP one but pretty damn effective nonetheless. I've never had the opportunity to work with the multi-$1000 ones like ORCAD, PADS, etc. Some of the PCB makers also offer their own packages (Sunstone offers PCB123 for free for example) but you accept the limitations that come with that if you decide to use such a thing.

No way I would use a single-purpose "tach" chip for this. All that does, is postpone the inevitable, when that one gets discontinued too. Not to mention their HORRIBLE non-linearity. Trigonometric functions are TOUGH to calculate fast and accurately at the same time with cheeeeeeep silicon.

ndndndnd

I did say I would be interested!

I'm dipping a toe in the water, by looking into this electronics stuff - it's very new to me. Looking on here: http://pcbshopper.com/ the cost of producing 'prototype' quantity boards doesn't seem too strong. Admittedly, it is more expensive if you go looking outside China and Hong Kong - but $6 per unit for five 3" square boards, or $3 per unit for ten isn't too scary. Admittedly, you know more about this and I doubt these boards meet quite your standards.

What would you have proposed regarding the PCB? Would you have retained the existing board and substituted new-for-old equivalent components?

sofakingdom

iTrader: (1)

Cool! That's ONE so far, then.

What I propose to do is to make a completely new PCB that replaces the old, using a microprocessor to drive the meter instead of a definite-purpose chip. I think (not completely sure yet) that for Camaro at least, which are the ones I'm familiar with, that I can make one board with all the components that will plug into the place where the later-model version plugs into the cluster backplane; and then make an adapter board to replace the one that the early-model uses, which is much larger and bolts to the back of the meter, with a plug to plug the other into. Cheeeeeeeeper to do it that way than a complete other design.

Here's the early-model board. Some of the holes in it are where the meter attaches, some are where it attaches to the cluster, the screws are where the flex circuit plugs onto it.



Buying PCBs out of China is REAL scary. There's NO TELLING what you're going to get, or when, or what you're going to do if something goes wrong. Been there, done that. No way I would go that way no matter how cheeeeep they say it is. A European or US supplier would be much more predictable.

The attachment is the later-model version that Amillionoh7 sent me. Note that the components on the 2 are the same; they only differ mechanically.

OrangeBird

iTrader: (1)

Sofa , I totally understand the "economy of scale" with regards to price and if I were to trust someone to build me a replacement board I couldn't think of anyone I'd rather see build it than you . Of course my car is a Firebird so I'm one step away from your planned Camaro boards and it's just but a wee tiny 6 and not an 8 which knocks my board down one more step away from what you want to create . So now in effect I'd need a submodel of a submodel .

Right now , my tach appears accurate , but that notwithstanding , If I could buy a home manufactured board of better quality than the original for anything around $50 to $75 , I would buy it as an upgrade just for the piece of mind of having the better components and thus being less likely to fail .

sofakingdom

iTrader: (1)

I would include some mechanism for switching the # of cyls; so that's not an obstacle.

For the Bird people, what I really need is samples. They don't have to be good, or even work at aall. The primary info would be dimensions, connector pinouts, etc.

loopy

iTrader: (1)

Sofa. I would buy one but can't send old one back economically from New Zealand.
Other than that I am keen

My tachometer reads way too high.

ndndndnd

$50-75 sounds perfectly reasonable - that's cheaper (and less risky) than getting a second hand set of clocks from eBay, and cheaper than getting some PCBs burnt, trying to source all the components and trying to learn electronics myself!

How would the new PCB 'plug' into the existing 82-89 tachometer board? Would the existing oil gauge be retained on the existing board as it is? Would connecting the new tachometer board be something I could do myself? It would also be impractical for me to post my tach from the UK...

sofakingdom

iTrader: (1)

There would be a board that would replace the large one, with a plug for the small one. All other functionality would be retained.

sofakingdom

iTrader: (1)

OK guys, update:

I've researched the various tachs that came in these cars a little bit. Pretty sure there are more than what I've seen so far. Seems that they are all built on the sine/cosine coil moving magnet principle, which is helpful: what that type of meter does, is to allow a specific pointer position to be commanded, but in the analog domain, not digital (i.e. it's not a stepper motor).

The fact that they all work on the same principle helps in that only one type of driver circuit would be required.

Unfortunately however, that's where the commonality ends. Some Camaros have a 180° range from zero to full-scale, some have a 270° range. Firebirds as far as I can tell all have a 240° range but I don't pretend to know all of their variations. Some cars are 6000 RPM full-scale, some are 7000, I've heard of but haven't actually seen that some are 8000 RPM. Engines can be either 4, 6, or 8 cylinders.

So, electrically, there are 27 potential electrical variations so far, that I know of:

4, 6, 8 cyl
180°, 240°, 270° deflection
6000, 7000, 8000 RPM full-scale

3 × 3 × 3 = 27

Then, just in Camaro, there are at least 2 mechanical versions, show in the above photos. Multiply again...

Without more information, can't say how many mechanical variants there are in Firebirds, but I'd expect the number to be similar or greater.

Bottom line is, this is NOT a one-solution-fits-all situation.

Good news is, it might be possible in some, most, or potentially even ALL cases, to adapt the smallest form-factor to fit into the space and connection constraints of the larger ones. As hinted at above, this would be done by either making a new "base" board to replace the existing one, and just putting a plug on it to accommodate the "universal" electronics; or possibly, adapting (some combination of drilling, cutting traces, soldering in wires, adding a connector, etc.) to the existing "base" board.

Cost it looking like it will be in the $50 - $75 range for the electronics, as guessed at above; probably closer to $75 than $50. Looks like this board would be a direct plug-in to the 90-92 Camaro, that being the smallest one. All other applications would then require some form of adaptation. No idea yet what the adaptation would cost; probably somewhat less than the actual electronics, but until I see more of the variants, I can't be very specific. Let's just put a WAG on it and say $35 - $50.

The electronics would have different software (well, actually the same software, just different constants programmed in) for the different electrical variations. IOW, they would have to be made to order for each application (best way of differentiating would be to just send me a photo of the dash you plan to put it on and tell me how many cyls), but could be changed later. I thought I could use jumpers or switches or something on the PCB to do that in the field, but there's absolutely not enough room for that, which is kind of a bummer.

What say you?

Amillionoh7

iTrader: (1)

sold my Camaro so don't need my tach back. you can sell it if u fix it

OrangeBird

iTrader: (1)

First thing I want to say is , Thank You Sofa , for taking the time to do the legwork on determining how this would work .

I can see with so many variations that a one size fits all solution won't be possible here and I can see where that would drive the cost up , having to build the different units rather than to be able to do a run of say 50 or 100 all of the same unit . I think , and this is just my own personal opinion , that folks will balk as the price approaches $100 , and I can't see it being at all profitable for much less than that . Your time is valuable as you know and you do deserve to be fairly compensated for that which along with parts I can see where $100 or darned close to could very well be whats fair . I think the disconnect between whats fair and what folks will actually pay may be the end of a great idea though . This thread has now been up one day short of 2 months and very few of us have responded when a $50 to $75 pricepoint was being discussed , at $100 I fear the response would be , unfortunately , resoundingly underwhelming

sofakingdom

iTrader: (1)

I'll have a few circuit boards in a week or 2; don't know when I'll get the time to assemble them, probably another week. I've got started on the software but until I have hardware to put it in it's hard to say how long it will take to get that into a usable state.

Here's an image of what the board will look like. The main dimension is 2.625" x 1.125", plus the connector part extends .250" beyond that. To give you an idea of the scale of things, the connector fingers at the top, which plug into a card-edge connector, are on .100" centers.

This is the top side; there's a few parts on the bottom too.

bigarmzz

if you would happen to have a tach for a 92 RS tbi i'd be willing to buy it

sofakingdom

iTrader: (1)

I'm thinking I will, that in fact, that application will be the first one I'll be able to fill. I'll post up on here when they're ready to ship and we can talk about it then.

Orange, unfortunately you're probably right. The economic reality is unpleasant sometimes, and people talk ALOT more than they're willing to walk. Especially when it comes to ACTUALLY buying the otherwise unavailable things they say they want when they become available but aren't mass-market low-bid low-budget chinese high-volume bottom-feeder CRAP {*cough*SkipWhite*cough}. I can't complete with that sort of thing, and for that matter, don't want to. If that means a bunch of cheeeeeeepsters drive around with broken tachs bitching about "why don't 'they' make a 'reasonable'...", .... so be it. It's their problem not mine. You can't accuse me of not at least putting forth the effort on their behalf.

But we'll see; this might be one of those things that doesn't sell all at once, but will be a one here one there kind of deal. I'll keep the price as low as I can, but there's only so much money I can afford to give away. I have no doubt that at least a few TRUE enthusiasts will want to do the right thing for their cars.

OrangeBird

iTrader: (1)

Well since your going for it , I too hope this will be one of those "if you build it , they will come" situations . Like I said earlier I will be a willing customer for one , when you get to the point when your 8 cylinder ones are all done and you expand the line to include 6 cylinder ones . My tach does work fine at the moment but I'm behind this idea enough that a better board is worth the money to me . I will of course be following this thread to monitor your progress and I will enclose a photo of my tach in case it helps you in any way .

sofakingdom

iTrader: (1)

That looks pretty much like most of the Bird ones I've seen... 240 degrees. (although I'd like a more precise measurement)

loopy

iTrader: (1)

When you can nail down a price LMK, Mines all screwy and I'm keen for one for a 83 camaro......... price dependant