I agree with Kev. It may be noted that the purpose of the relay is twofold:

- to keep the lights brighter by lessening the length of wire
- to reduce wear on the switch or to be able to use a smaller switch.

If you used a heavy duty switch, it may work fine electrically. But should still be fixed.

Quote:

That's true, and not a bad idea. I've used Ardunios many times in the past. And an Arduino Nano has the necessary ADC on it to process an input of variable resistance. The main drawbacks are that it can't be standalone. It needs:
- power supply (6-15V to 5V)
- precision voltage divider for the input
- output driver, since the 20mA per pin of the Arduino is only about 1/25 of what's needed to drive the gauge reading.

Also, while programming these is simple enough, it could be a benefit to sell something the user can just set up and trim themselves. Of course you can add some pots to the Arduino but at that point you may as well make a custom daughterboard. And then we should consider an enclosure for it...

This really shouldn't be terribly hard with a few op amp circuits combined. Will have to play with it but have been under the weather today, so not been on my PC.

Thanks Howard. I'll read through those shortly.

I have a Chilton's (?) manual but idk if it goes back that far.

From the service letter and DP's data it's clear that the water temp gauge indicates backwards compared to the other two, but with the same range. At least for the gauge, this can be easily accomplished by physically reversing the assembly inside it during the manufacture process.

Therefore it may be assumed that the oil and gas gauges operate according to DP's data, just reversed. I.e. a high impedance at the sender gives a low gauge reading and vice versa.

It should be noted that all 3 gauges on the later cars followed this same pattern of high impedance = low reading. Actually I'm not certain on the oil pressure gauge, so maybe Howard can confirm that.

I think if there is a device that can trim both for an input threshold, and an output range, and can reverse the logic between input and output, it could be easily set up to make any sender work properly with any gauge. I'll have to see if I can come up with something. Of course, this would work not only for the 48-50, but the other obsolete types and also for all makes. Thr only assumption is that both scales are linear, so if I can make a decent "reverse exponential" device separate for 48-50 it may be best to do so.

Quote:

I appreciate it, though I'm not sure it would be needed. Since what I have in mind would be that someone can buy a generic sender with a linear response, and have that adapt as well as reasonable to the factory gauges.

The other thing is that, as far as I am aware, the usual rheostat senders have a low resistance when full and high when low. But David's chart is the opposite.

Edit: Actually seems like low ohm empty and high ohm full is the standard, at least for GM.

Thanks dp. I don't know, but would assume, that the gauges are electrically identical.

Howard or dp do we know for certain that all 3 senders used the same signaling scheme?

Thanks. It's surprisingly non-linear, but maybe that's how the gauges read to begin with.

Since you have 0.2A at 20 sender ohms, then with a 5V regulator that implies the gauge has a 10 ohm resistance. This checks out with the 50 and 100 ohm values also. So we see that about 22mA is required for the gauge to read 0; and open circuit is required for the gauge to read 100. I can easily make something that will convert linear-linear. Following the exponential decay curve could be difficult, but might not really be necessary as you can just calibrate the gauges to what looks good. For example, where normal coolant temp and oil pressure read right in the middle, and where the gas gauge reads perfect near empty. Will have to think more on a design that could replicate this curve well.

David please confirm, the "gauge percentage" of 0 is like your empty tank, no oil pressure, and cold engine; and "gauge percentage" 100 is full tank, high oil, and high temp?

P.S. Howard, I think that's what I'm getting at. I can make something where a 10-70 ohm sender will pretend to be a 20-120 ohm or whatever. Of course, if the range only needs shifted then a simple inline resistor would do the trick. But for these, even though a 10 ohm would make a 10-70 sender read correct when low, it would never read about more than about 85% high.

Good idea but I don't have a 48-50 car to try anything. So I'm not sure what to do with that info. Someone else would have to try it 😃

As many of you know, the 48-50 cars used unique senders (gas tank, temp, etc.) that are pretty much no longer available.

I can make a module that can electrically "convert" between any generic aftermarket sender that physically fits, and the corresponding gauge. It would only require a brief calibration.

But to do this, I would either need someone to send me a spare gauge, or be able to send some pictures plus maybe do a couple basic electrical tests on a gauge. (Not sender.)

Please let me know if you would be interested in this project. The pricing would be very competitive in comparison to alternatives.

Quote:

Certainly. And not often found on sites dedicated to other makes or models.

I actually complimented it on my second post ever here 😃

https://packardinfo.com/xoops/html/modules/newbb/viewtopic.php?topic_id=24406&post_id=235489#forumpost235489

Idk, parts book gives them all the same.

https://packardinfo.com/xoops/html/modules/partslist/viewgroup.php?partslist=1955-1956&groupnumber=4.1337&PlateNumber=31&cat=23#resultanchor