Aftermarket Gauge Install

[pommeree]

Gentlemen,
I humbly submit as case-study, my recent gauge install project.
I hope to enjoy (constructive) feedback from those more knowledgeable than I and to help inform those who wish to follow.
Please reserve comment until you've read it through as I've provided my own "observations" at the end.
Also, Images are sized based on necessary detail. If you want them resized, lemme know because thanks to picasaweb, I can do that in 5 Sec.
Also included at the end is a link to the album of all photos plus extras.

Anyhoo. Some time ago I purchased on e-bay the VDO oil pressure and temp gauges with corresponding senders as supplied by Bavauto.

Note on the back of the packages, VDO supplies Ohm specifications. These represent the lowest and highest readings on the corresponding gauge. (I came to this realization while on the phone with a VDO support rep who was providing low, middle and high specifications for my temp sender. Still troubleshooting. More on that later.)




My plan was to use VDO gauge pods mounted atop the faux speaker grill as I felt the black, minimalist look was in keeping with the (awesome) aesthetic of the exiting dash.
Here's the hardware as it ships from VDO (Summit)


Here's a quick preview of how it's going to look


The first step was figuring out how to assemble the gauges in the pods.
For the life of me I couldn't figure out how these were supposed to go together. I sincerely believe that VDO put wrong parts in because, in spite of supplying plenty of extras, there were nuts that weren't threaded to fit the corresponding bots and holes that weren't big enough to accommodate bolts, etc. What was supposed to be an "out of the box" solution instead required some fabrication work.
Here's the design I settled on.

I drilled out the tip of the "V" bracket to accommodate the retaining bolt (I replaced the stainless one they supplied with a black 10-24x1" allen bolt and corresponding nylock nut JB Welded into the V-bracket.




Who'd pass up an opportunity to use JB Weld?


Thanks VDO for making me drill ...


OK. Once I had all that bull &$% sorted out, I did a test mock up on the speaker grille. Again using 10-24 black allen head nuts and nylock washers.





Underneath I drilled two holes in some flat aluminum stock for reinforcement and slotted out part of the grille for wiring.
Important note: There is a "front" and a "back" to the speaker grille. Make sure you know which way is which before drilling and cutting.
(hence the "extra" wiring slot. Oops.)


OK. Now that things look real cool. It was time to wire it up and see if they'd actually work. I did some homework and found that there is a perfect source for power pre-wired under the dash.
The harness you see below is situated between the washer-interval device (black box on left) and the speedo cable where it comes into the cabin (on the right.) The green/white wire is the start-run bus (switched power) and the grey/blue wire is "dash lights".
If you have an automatic transmission, this may complicate things as I believe this harness supplies power and light to the Auto tranny selector.


But wait. It gets even better. Move the Washer interval box out of the way and behind it is a ground!


Boring wire pulling pics.
The only thing I'll add here is I estimated total wire length in each of three colors for two runs and grabbed in the middle for the pull.
In other words, six runs in one pull.
Also, a question: I used 16 gauge wire throughout. It seems a little heavy. Would a thinner gauge have sufficed?

Guesstimating ...


Tie them on ... (that's the end of a flexible magnetic retriever I used as a fish-tape (puller))


A better perspective ...


Voila! Six leads, one pull.


Don't ask me why, but I wanted to ground the lights separately.


For that I chose another convenient ground located to the right of center console. See the screw in the dead-center of the photo.


To run wires to the senders:
I gently shoved a long-handled screwdriver in between the grommet and main harness where it comes in by the glove-box. This produced a gap big enough to get the two, small lead wires through.

Here's a detail on the oil pressure sender:
Copper crush washers were $1.39 each from NAPA and I used permatex thread sealant on the sending unit.
The adapter-block came with the gauges but the 10x1mm threads for the sender were jacked up. Let me tell you this: Good &#$% luck finding a 10x1mm tap here in the US. Every store on the planet sells inexpensive tap sets and every one of them has a 10x1.25mm or 10x1.5mm tap.
Also, the lead for the pressure warning light won't reach the sender so I have yet to solder in an extension and hook it up.


For the temp sender, I cut the insulation off a female spade connector, crimped it onto the wire, slid heat-shrink insulation over the whole end, shrunk it and then cut a slit for where it slips onto the sender.
Hopefully that will keep it relatively weather-proofed.


Extra wire tethered to the engine mount.
Too close to the pipes? Uh ...


Hmm. That doesn't look quite as cool as I was hoping.
Lighter gauge wires? Separate holes for separate gauges?


This looks pretty good ...


From the drivers perspective.
I need to dress up those wires better.


As of today (8/19/09) I've put maybe 250 miles on the setup so far and here are my observations:

1. There is nothing securing the faux speaker grill to the dash and the gauges with pods are pretty heavy. One good pothole and the whole assembly can jump. Or worse, a fender-bender would send it through the windshield. I'm going to secure the speaker grille with some hidden/black screws.

2. The gauge pods, while they look awesome have two major flaws. A. They don't ship with the right &$% hardware and B. The slender steel brackets allow the gauges to "bobble-head" with vibrations in the car. This makes them look totally cheap. I'm going to install two machine screws from up-under the grille to meet the gauges at their chins to prevent them from &#$% bobble-heading.

3. While my pressure gauge worked perfect the 1st time, my temp gauge has a hard time breaking the Min. 120 Deg. mark. I get to start taking resistance readings from the sender and maybe grounding the lead with resistors in line to simulate temp readings to get to the bottom of that.

4. Theory: Every job has a fixed cost no matter which way you come at it. I got a deal on the gauges and senders on E-bay, but because I purchased them years ago, I had no recourse when I discovered the bad threads on the adapter and there may be issues with the temp sender or gauge. I got to spend $40 for a nice little Craftsman tap set to fix the threads and the jury is out on the temp gauge. If I had to do it over again, I'd wait to buy the hardware until I'm ready to install and test. I'd might even pay retail from Bavauto or the like because if there's any problems, you can just send it back rather than having to improvise.

Here's the linke to all photos:
Link to all Photos

That's it. I look forward to your comments.

[gabmwdrvr]

Nice write-up. I had to chuckle when I saw the hazard button. Check the FAQ's on how to repair it. It is very easy.

[2002jm]

Nice job, and great pics!!!! Maybe put some of that black wire covering stuff on the wires coming out the speaker cover.
jm
Oh, I now see you have some type of black cover on the wires, maybe bigger size will make it look even better.

[DEZIAC]

cool, nice write up, i'm bookmarking this for when i get time to install my gauges. Still need to get the sending units. Anyhow, great work!

[1980 320iS]

Good write-up - thx for posting the info.

If it was me, I would extend the same black covering for the wires through the faux speaker grill (ie, until the wires are out of sight). Let us know what you come up with to secure the speaker grill...

Oh, it looks like your bike is about the same vintage as your E21 -- glad I'm not the only one with that scenario!

[pommeree]

Originally Posted by gabmwdrvr

Nice write-up. I had to chuckle when I saw the hazard button. Check the FAQ's on how to repair it. It is very easy.

I've lost track of how many times I spent swerving down the freeway trying to jam a matchstick or gum wrapper into the thing to turn my Hazards off.
Or waking up in the morning only to discover they came on some time in the middle of the night.
I've repaired at least two. After I nearly sliced my finger off trying to pry one apart (for a hazard switch I'd never needed, before or since) I though, "&#$% this. I'm fixing it once and for all."
Two eyelets and a cotter-pin. Done.

[DrewDude320i]

Don't drill anything. If you want, we have a TON of 3M double sided foam tape and VHB tape at work just sitting in a cabinet, left over from a huge sample roll they sent us. It's the same stuff used to hold the glass onto the metal frame in the Burj Al-Arab Hotel in Dubai, so it should be able to handle your plastic grill piece. If you measure how much you need, I can cut you some.

[pommeree]

Originally Posted by 1980 320iS

Oh, it looks like your bike is about the same vintage as your E21 -- glad I'm not the only one with that scenario!

Frame: 1984 Austro-Daimler 'Ultima-SL' by Puch. Reynolds 531 for everything, including the fork. Campy headset, Bottom Bracket and drop-outs.
Last time I rode it? Uh ... 1980-something?

[sonomaGTLN2]

Looks good...looking forward to your tweak ideas.

[gabmwdrvr]

Just saw the bike. How tall are you? That thing looks like 60cm or more and the seat is raised real high. I love vintage bikes!

[superj]

i think they look great man. when i do some, they are going in the center vent slots since i have no ac anyways and only use the defrost setting.

[pommeree]

Here' my arrangement for testing the sender.
The circuit is:
Wire from sender through dash to Multimeter(+)
Mulltimeter(-) clipped to ground under dash.
We can eliminate the wire as a problem b/c resistance is identical in this arrangement as it is when measuring the sender directly.


My first questions are these:
The resistance appears to be greatly influenced by other electrical events in the car. Is this normal for a sender lead? Is the gauge influenced the same way and therefore compensating in some way? Do I need to figure out how to isolate my sender circuit?
Examples:
Turn cabin fan on Max. -20 Ohms.
Turn on Aux Rad. Fan -20 Ohms.
Turn engine off +100 Ohms.

Example: in the photo you can see ... with the engine off and warm, I get ~138.0 Ohms. (same as at the sender)
Before I cut the engine off, it read ~38. Ohms.

(This is particularly confounding because, not having yet tested the gauge independantly, 38 Ohms should give a reading somewhere between 260-300. But my gauge has never made it past 200.)
Temp. Gauge testing forthcoming as soon as I hunt down my breadboard and collection of resistors.

[1980 320iS]

One thing to try: Instead of having multimeter(-) clipped to the ground under the dash, connect to a ground near the sender. (Your DVM leads appear to be juuuuuuust long enough to hang the meter outside the window and have the neg lead reach ground points in the engine bay.)

If you turn on your fan(s) and such and the resistance does not change, then all you have to do is run a separate ground lead for this gauge and you're done. (Kinda sucks to do again what you've already done but...)

From what you've described, I'm 95% confident that you've got a ground loop. (As I often say, not all grounds are created equal.)

[superj]

it sounds like the engine is trying to ground through your gauge wires. put a ground wire off a bolt that is really close to your sender or something like that. you might just not have enough grounds on the engine. i had to put a few extra on my e30 and then it ran great

[GDAus]

Originally Posted by pommeree

Here' my arrangement for testing the sender.
The circuit is:
Wire from sender through dash to Multimeter(+)
Mulltimeter(-) clipped to ground under dash.
We can eliminate the wire as a problem b/c resistance is identical in this arrangement as it is when measuring the sender directly.

I deleted the photo to help preserve the site's bandwidth !!

My first questions are these:
The resistance appears to be greatly influenced by other electrical events in the car. Is this normal for a sender lead? Is the gauge influenced the same way and therefore compensating in some way? Do I need to figure out how to isolate my sender circuit?
Examples:
Turn cabin fan on Max. -20 Ohms.
Turn on Aux Rad. Fan -20 Ohms.
Turn engine off +100 Ohms.

Example: in the photo you can see ... with the engine off and warm, I get ~138.0 Ohms. (same as at the sender)
Before I cut the engine off, it read ~38. Ohms.

(This is particularly confounding because, not having yet tested the gauge independantly, 38 Ohms should give a reading somewhere between 260-300. But my gauge has never made it past 200.)
Temp. Gauge testing forthcoming as soon as I hunt down my breadboard and collection of resistors.

Ok .. A couple of things ...

Firstly ... Using a multimeter to measure ohms like this is NOT VALID! You need to disconnect the car's -ve battery connection before you measure to get a valid resistance measurement.
This is because the way the ohms function works inside the multimeter is to pass a current through the circuit using the MULTIMETER'S INTERNAL 9 volt BATTERY and the circuitry measures the voltage drop across the circuit and gives you a reading in Ohms. The fact that you are getting a -ve ohms reading shows that the car's electrical system is pushing current through your multimeter in the opposite direction, overwhelming the meters internal battery. (there is no such thing as negative resistance ... unless you get into the realms of superconductivity lol .. note: negative resistivity is something else )

The + 100 ohms reading may be somewhere near what the sender unit was at that point, but the only way to be certain is to disconnect the battery if you want to measure resistance directly .... but a more useful result for this purpose is to measure the voltage drop and current across the temp sender and deriving the resistance that way.
BTW ... all electrical gauges are measuring the voltage drop across their respective sender units and connecting wiring ... the faces of the gauges are simply marked to indicate the equivalent value .. ie temp etc. There will be a slight difference in the actual reading of the gauge when it is in circuit between engine off and running, because with the engine running the charging system should be keeping the electrical system at around ~13.8 Volts. Some cars have voltage stabilisers in the instrument wiring to compensate but some don't (like the E21's ..lol) so if you want to be precise ... measurements should be done with the engine running fast enough to give you a stable system voltage.

So ... with the meter set to voltage, connect your meter's -ve lead to the wire FROM the temp sender and the +ve meter lead to a +ve supply ... record this voltage. Disconnect one meter lead, change multimeter to read amps .... reconnect lead ... record current. Calculate resistance from R=Voltage/Current. This will give you a real world result based on the actual system voltage in your car.

The reference to possible ground loops ... this is more of a problem with audio and data connections ... not so much a power problem unless there are some seriously bad connections in the ground points that are being used.

Cheers

[paule21coupe]

thanks for taking the time to post this thread, healpfull..

[DrewDude320i]

Alright, problem.

I decided to re-wire my gages (Wideband O2, Boost, and Temp) to the plug you found under the dash. The boost gage's background light is hooked up to the blue wire, and the )2 sensor gage, temperature sensor and temperature gage are hooked up to the greens. When my lights are off and the key is in the "on" position, everything is on, but:

1) my O2 sensor keeps resetting, like it's not getting enough power (didn't happen before)
2) Whenever I turn on the parking lights or the headlights, all the gages lose power.
3) If I swap the connections (blue to green and green to blue), then just the boost gage light (which is on its own power wire) comes on, but again turns off when you turn on the parking/headlights.
4) Unplug any of the two wires, and all gages lose power.

Any idea what's going on?

[GDAus]

Originally Posted by DrewDude320i

Alright, problem.

I decided to re-wire my gages (Wideband O2, Boost, and Temp) to the plug you found under the dash. The boost gage's background light is hooked up to the blue wire, and the )2 sensor gage, temperature sensor and temperature gage are hooked up to the greens. When my lights are off and the key is in the "on" position, everything is on, but:

1) my O2 sensor keeps resetting, like it's not getting enough power (didn't happen before)
2) Whenever I turn on the parking lights or the headlights, all the gages lose power.
3) If I swap the connections (blue to green and green to blue), then just the boost gage light (which is on its own power wire) comes on, but again turns off when you turn on the parking/headlights.
4) Unplug any of the two wires, and all gages lose power.

Any idea what's going on?

Really need a bit more info on what type/brand gauges your using .... ie, I guessing your using a wideband O2 controller gauge ... but is your Boost gauge a controller too? ... digital or analog ..etc etc? ... and where are you taking the -ve supply from ? (current flow in a D.C. circuit is actually from -ve to +ve, so the ground connection is the most important!) a diagram of your connections would be useful

Also ... the Grey/Blue wire is from the instrument light dimmer rheostat ... ie, depending on the setting, it will be less than +ve 12 volts and is only "on" when the lights are on. (and ignition depending on model) The rheostat will also cause a certain amount of "noise" in this circuit and shouldn't be connected to anything like a wideband controller. In fact, a wideband controller should have the cleanest practical power circuit ... it's own -ve cable right back to the battery will help keep it accurate and stable.

BTW .. I have to ask .... If all this was hooked up and working .... why change it ?


Cheers

[DrewDude320i]

Originally Posted by GDAus

Really need a bit more info on what type/brand gauges your using .... ie, I guessing your using a wideband O2 controller gauge ... but is your Boost gauge a controller too? ... digital or analog ..etc etc? ... and where are you taking the -ve supply from ? (current flow in a D.C. circuit is actually from -ve to +ve, so the ground connection is the most important!) a diagram of your connections would be useful

Also ... the Grey/Blue wire is from the instrument light dimmer rheostat ... ie, depending on the setting, it will be less than +ve 12 volts and is only "on" when the lights are on. (and ignition depending on model) The rheostat will also cause a certain amount of "noise" in this circuit and shouldn't be connected to anything like a wideband controller. In fact, a wideband controller should have the cleanest practical power circuit ... it's own -ve cable right back to the battery will help keep it accurate and stable.

BTW .. I have to ask .... If all this was hooked up and working .... why change it ?


Cheers

I'm using an AEM Wideband O2 gage, which is a controller. The Boost gage just has a light, not controlling anything. The temperature gage has a little temperature sender that it powers. The boost and temp gages are digital.

Only the O2 gage was working before, and it was wired straight to the battery. I hate that - I want it to look a little cleaner. I have enough of a rat's nest in my hood as it is.

[GDAus]

Originally Posted by DrewDude320i

I'm using an AEM Wideband O2 gage, which is a controller. The Boost gage just has a light, not controlling anything. The temperature gage has a little temperature sender that it powers. The boost and temp gages are digital.

Ok ... the brand(s) of your boost and temp gauges would have been helpful, but I can make some reasonable assumptions and I think I know what you have done.

Firstly, since all the gauges have digital readouts, they all need a 12 Volt +ve & -ve supply to power the electronic readouts etc. (and yes, the UEGO controller also supplies the heater in the wideband O2 sensor). So, all the gauges should have a Red and Black lead (and others as the case may be), which is +ve and -ve respectively. I'm assuming you have connected all of the reds together ... and all of the blacks together which is all good so far

However, to give the symptoms you are describing, it sounds like you have connected the Reds (+ve) to the GREEN/White connector (which is good .. it is a switched + 12 volt supply from the Start/Run Bus) ... BUT if you have connected the Blacks (-ve) to the GREY/Blue connector, then the stuff you describe could happen

If this is the case ..... then this would allow the gauges to work (sort of) while the lights are off ... because there would be a current path from -ve ground through the filaments of the instrument lamps which are all connected in parallel. But there would be a voltage drop caused by the resistance of the lamp filaments so there would be less than the usual system +ve voltage across your gauges and that would match up with your observation of your .... "O2 sensor wasn't getting enough power" so it was resetting.

Now, when the lights are switched on .... the Grey/Blue connector would be at close to +ve 12 volts (depending on the dimmer setting in the light switch) so there would now be +ve voltages on both the GREEN/White and the GREY/Blue connectors ... hence all the gauges would shut down !

So .. Just to be clear .... you should have :
REDS connected to the GREEN/White connector
Blacks connected to a -ve chassis ground point (best would be the main ground point under the dash pictured by Pommeree)

That should get your gauges working

Not sure about the Blue lighting wire from your boost gauge ... I would have thought a digital gauge wouldn't need an internal light ?? or is it a sensing lead that dims the readout at night ? If that's the case, then connecting that to the GREY/Blue connector is the way to go.

Hopefully this will get things sorted

cheers

[DrewDude320i]

Sorry, I should have been more clear - my boost gage is not digital. It has a background light to see the needle at night.

I'll double check my wiring, but I'm fairly certain I have the +ve wires from the O2 gage and the temperature sensor hooked into the green/white connector, and grounded off behind that connector - and the "night light" for the boost gage connected to the blue /grey wire and grounded off in the same spot as the others. If that's the case, I can't figure out why ALL of them would turn off when the lights come on, especially since the boost gage light is only supposed to be on when the headlights are on.

[pommeree]

Originally Posted by GDAus

The fact that you are getting a -ve ohms reading shows that the car's electrical system is pushing current through your multimeter in the opposite direction, overwhelming the meters internal battery. (there is no such thing as negative resistance ... unless you get into the realms of superconductivity lol .. note: negative resistivity is something else )

The + 100 ohms reading may be somewhere near what the sender unit was at that point, but the only way to be certain is to disconnect the battery if you want to measure resistance directly .... but a more useful result for this purpose is to measure the voltage drop and current across the temp sender and deriving the resistance that way.
Cheers

I am an idiot for not having written more clearly.
At no point was I getting a negative Ohm reading.
What I meant to express was the relative impact of each of the devices.

I'm going to run a ground wire into the engine bay and test. I'll have more readings after the next drive.

[pommeree]

The good news is, having run a new ground wire to the block, my tests of the sender are far more consistent. There is no longer any influence of other electrical devices on the resistance readings.

The bad news is, my readings are consistently *wrong*
On startup (cold) I had 915 K-Ohms.
Highway (warmed up) I had 180 Ohms.
City (warmed up) I had 130 Ohms.

This points to a bad sender. However, if I ever find my electronics kit (w/ a little breadboard in it) I hope to ground-out the sender lead with resistors in-line to verify that the gauge itself is working as intended. I'll be sure to post pics of that.

Also going to rework aesthetics of the wiring through speaker grill. The whole project was built around the concept of a sacrificial speaker grill and in retrospect, I was being far too anal about sacrificing it.

Hey DrewDude320i,
How are you making out?

[pommeree]

Beta trials are done. Only real hiccup is a bad temp sender so I'll swap that at the next oil change (or maybe the one after that ...)
Anyway,
Rats nest under the grill had to go.


Make a "harness" out of the wires using old school friction tape.


Medium picture. You get the idea.


No more rat's nest under the grille.


Ready to be trimmed, crimped and connected.


Modified the grille with "anti-bobble" posts threaded directly into the grille.


Tidy wires, bobble-posts and screws with dress collars (formerly chrome, painted flat Blk.)


Ready for final dress ...


Done


Done 2



Some parting thoughts:
1. I'm undecided on the aesthetic of the wire rap. Maybe just need to take out the extra slack. (Made install easy, though)

2. The angle of the gauges makes for pleasing symmetry of elements on the dash ... and headlights shining in my eyes at night. Will need to tweak the angle a bit ...

3. Even with the dash vents closed, cold gauges and warm morning fog make for a little condensation inside the gauges. Hasn't been serious yet ...

4. Gauges that glow red and dim with the dash lights is über-cool. But now my main lights look too dim. Get to go after them next.

[aymaym]

good work