Hi Alec_t it works perfectly So happy! I added an red LED to make it more boy racerish
Took me ages to find anywhere that would do just a few resistors etc, In the end i remembered Tandy, and they did everything for under £5 posted!

 

Thanks for the update. Good to know it works in the real world .

 

Thanks for the update. Good to know it works in the real world .

Hi Alec
I have a bit of an issue, when the engine fan cuts in it seems to have blown my little interface as the dash no longer detects anything. I guess the op amp has blown, so I have orders some more. Is there anyway I can add protection?
regards nick

 

Unfortunate that, tho the op-amp is good to 32v so a switching spike shouldn't have hurt it but who knows

Add two 15v zener diodes as shown at D2/D3 should help next time!

 

Unfortunate that, tho the op-amp is good to 32v so a switching spike shouldn't have hurt it but who knows

Add two 15v zener diodes as shown at D2/D3 should help next time!

View attachment 212367

 

is there anything else that could have blown? The dash is not seeng anything.
could it be the transistor?
Is there another part number for the diode, all the ones on the Tandy website have numbers like 1N4744A 15V Zener Diode
kind regards

Unfortunate that, tho the op-amp is good to 32v so a switching spike shouldn't have hurt it but who knows

Add two 15v zener diodes as shown at D2/D3 should help next time!

View attachment 212367

 

is there anything else that could have blown? The dash is not seeng anything.
could it be the transistor?

The transistor (more likely) or D1 (less likely) are possibilities.
The 1N4744A will be fine.

 

The transistor (more likely) or D1 (less likely) are possibilities.
The 1N4744A will be fine.

Thank you!

 

Yeah, I didn't look too closely for specifics. The transistor is unlikely, its good for 35-45v depending on manufacturer. I think the problem was exceeding opamp common mode input.. the opamp +rail is protected by D1 and R1C1 time constant, but the fuel gauge potentially is returned to +12v which could see large + or - switching spikes & maybe the opamp +/- inputs were pulled substantially above V+ and/or below zero.

 

That seems a likely explanation. I had assumed the instrument cluster would include voltage regulation and protection components around the gauge. I guess the more protection the better when it comes to auto electrics.

 

Measure the voltage between the + and - terminals of the OP-amp. It should be near zero.

You can also put nearly any value of resistor between 10K and 1M in series with the + input of the OP-amp and also the - input of the OP Amp.

You can put a TVS diode https://www.digikey.com/product-detail/en/on-semiconductor/NVR4003NT3G/NVR4003NT3GOSCT-ND/5969169 from +12 to ground (band to +12).

and put an 1n400x, x>=3, e.g. 1N4003 on the +12 and ground with the band on the +12 side.

Put a 0.1uF ceramic capacitor between +12 and ground.

Also add a 0.1 uF ceramic capacitor (bypass capacitor) close to the OP amp

The schematic doesn;t show the power supply pins on the LT1077. Pin 4 should be ground and pin 7 to +12 with a 0.1 ceramic capacitor to ground very close to the IC.

Note: fixed TVS polarity.

 

Some good points there. I only copied the original circuit to show where the zeners should go. I'd put a 1nF and a 100nF ceramic in parallel from the opamp pin 7 to ground (opamp pin 4) to trap high speed pulses.

I might consider putting 10nF to ground on sender input and across R2 and add 2 x 10k on inputs to op-amp.

That seems a likely explanation. I had assumed the instrument cluster would include voltage regulation and protection components around the gauge. I guess the more protection the better when it comes to auto electrics.

It probably does, but those are potentially long wires in a noisy environment. Auto electrics are a very hostile environment.

 

Measure the voltage between the + and - terminals of the OP-amp. It should be near zero.

You can also put nearly any value of resistor between 10K and 1M in series with the + input of the OP-amp and also the - input of the OP Amp.

You can put a TVS diode https://www.digikey.com/product-detail/en/on-semiconductor/NVR4003NT3G/NVR4003NT3GOSCT-ND/5969169 from +12 to ground (band to +12).

and put an 1n400x, x>=3, e.g. 1N4003 on the +12 and ground with the band on the +12 side.

Put a 0.1uF ceramic capacitor between +12 and ground.

Also add a 0.1 uF ceramic capacitor (bypass capacitor) close to the OP amp

The schematic doesn;t show the power supply pins on the LT1077. Pin 4 should be ground and pin 7 to +12 with a 0.1 ceramic capacitor to ground very close to the IC.

Note: fixed TVS polarity.

Some good points there. I only copied the original circuit to show where the zeners should go. I'd put a 1nF and a 100nF ceramic in parallel from the opamp pin 7 to ground (opamp pin 4) to trap high speed pulses.

I might consider putting 10nF to ground on sender input and across R2 and add 2 x 10k on inputs to op-amp.


It probably does, but those are potentially long wires in a noisy environment. Auto electrics are a very hostile environment.

This is beyond me I’ll add the 2 zds and change the other components and fingers x’d !! Was alright until I put it i a little abs box!

 

I've analyzed a car manufacturer's audio amplifier to figure out how/why they die. They tend to die when the car gets jumped.
I do know that one turn-on transistor obviously self-destructed in one amplifer. I replaced it and it broke again when the car was jumped.

I'm anxious to be able to put these amps on the bench and actually find out what the issues are. Usually, the amps start to develop a thermal problem where they work below a certain temperature until they die altogether. I have one that has a single channel out which is an unusual failure. So, every time the car is jumped, or started or the radio is turned off, I suspect damage occurs a little at a time. It's just difficult to wire up the amp on the bench. The only troubleshooting I've done is in the vehicle.

I probably have like six of them. One works currently. I won't put it in the vehicle until I understand the failures and make the fixes.

To give you an idea as to how bad the automotive environment is, read: https://m.littelfuse.com/~/media/el...n_automotive_environment_application_note.pdf

 

To give you an idea as to how bad the automotive environment is, read: https://m.littelfuse.com/~/media/el...n_automotive_environment_application_note.pdf

Good article. Especially the notes about inductive switching transients and mutual coupling in harness wiring.

Back in the late 1970's I was a very junior design engineer on new-fangled battlefield digital radio communications (the forerunners of the cell phone we know today, but thats another story) and for testing these were fitted into C1950's Landrovers and Commer and Bedford trucks that had never been fitted with any suppressive equipment. Suffice to say, failures were rife and much head-scratching ensued as to how to replicate this on the bench. Eventually, it took a year or so, we got on top of the problem, as every new platform (boats, airplanes, etc.) brought a whole new set of issues. I learnt a lot, empirically, about suppression and designing for harsh environments in that year (temperature and cooling, or lack of both, being the other killer).

 

The amp is mostly these AN7190NK https://industrial.panasonic.com/content/data/SC/ds/ds4/AN7190NK_E_discon.pdf IC's which are pretty robust it seems. The schematic does not protect against positive transients in both the remote and power. negative transients are covered.

A transistor seeing 470 + 20K to the base of a transistor visibly cracked. No positive voltage protection. So, that nails one problem.

Mute and Standby can exceed 5V in their design and I don;t think it should, so two problems.

 

OK, I'll admit stupidity I changed all the components and it still didn't work, my car dash kept recording 510 ohms, (default for no connection/open circuit) when I used an AVO on the board I could see about 300 ish.. I checked all the strips for any bits of solder, nothing, then it worked, then it didn't.. loose *ing gauge wire

 

Fractured/loose wires caused by vibration can be a problem in vehicles. Make sure cable ends have mechanical anchorages to stop them flexing.

 

Fractured/loose wires caused by vibration can be a problem in vehicles. Make sure cable ends have mechanical anchorages to stop them flexing.

On a positive note, the circuit is now protected with 2 zener diodes and I have loads more components for my box

 

Here you go. I've updated the circuit to cater for your resistance values :-

View attachment 210246

Edit:
Note that this adapter is for a conventional 'old'school' analogue gauge, not a fancy modern digital one.

Hello Alec_t, I just saw this topic and I have the same problem but, in my case, the cluster waits a sensor from 4 up to 79 ohms and my fuel level sender has 40 up to 280 ohms... Could you please help me with the resistors values to be applied into your circuit? Thanks!