Author |
Message |
LandSailor Samba Member
Joined: October 30, 2009 Posts: 315 Location: Las Vegas, NV
|
Posted: Mon Jul 28, 2014 1:43 am Post subject: Help with blinking, Hoping to Leave on Tuesday |
|
|
Sorry if this sounds like I don't know what I'm doing, but after six solid hours messing with this with an eight hour work day before that, I can't even see straight. I've posted in the forum trying to get this figured out.
Please let me know if any of these conclusions are wrong.
I pulled relay 42 from the left side above the ground tree. The gauge doesn't do anything. That means the problem is between the Instrument Panel Connector and the gauge. My flex PCB is actually in pretty good shape. The only possible issue is with the circuit on the bottom: #14, iirc but that's not too bad. The three nuts for the gauge are in place. I haven't removed them to check for oxidation, but it doesn't sound like a problem.
What do I need to beep out? What do I need to check?
I'm an entertainment lighting and automation guy, so I can do this, I just don't know what to do right now. Please provide links, if possible, and I will check before I leave for work in five hours...after I shower and sleep.
Voltage reg???
On the upside, I did get that fresh air vent fixed! |
|
Back to top |
|
|
dbeierl Samba Member
Joined: April 22, 2009 Posts: 164 Location: Providence, Rhode Island, USA
|
Posted: Mon Jul 28, 2014 3:39 am Post subject: Re: Help with blinking, Hoping to Leave on Tuesday |
|
|
LandSailor wrote: |
I pulled relay 42 from the left side above the ground tree. |
So you have a 1.9l. Tall or short controller?
Quote: |
The gauge doesn't do anything. |
Honestly you have to describe what you did and what conditions were before and after a lot more specifically than this to get a definitive answer. I know you were tired.
Making some assumptions though: If you unplug the controller, cycle the ignition, and get a blinking light that doesn't stop, there are only two possibilities, and the likely one is it's an internal gauge problem, see my numerous posts in the rest of this thread. The other possibility is that the gauge isn't being fed ten volts (+/- 0.5V), check on the back of the gauge btw ground pin (the skinny one) and the higher-reading of the other two. |
|
Back to top |
|
|
Ahwahnee Samba Member
Joined: June 05, 2010 Posts: 9797 Location: Mt Lemmon, AZ
|
Posted: Mon Jul 28, 2014 5:37 am Post subject: |
|
|
My gauge would peg and the blinking light would flash due to a break in the wire from the pressure tank sensor. Took awhile to find it -- the break was at a clamp or tight turn the wire made somewhere under the coil.
But as I noted in your other thread -- doesn't seem like much point in chasing the gauge behavior when you have coolant issues. If you are losing coolant I don't know how you expect you can make the gauge act normal. |
|
Back to top |
|
|
dbeierl Samba Member
Joined: April 22, 2009 Posts: 164 Location: Providence, Rhode Island, USA
|
Posted: Mon Jul 28, 2014 6:39 am Post subject: |
|
|
Ahwahnee wrote: |
My gauge would peg and the blinking light would flash due to a break in the wire from the pressure tank sensor. Took awhile to find it -- the break was at a clamp or tight turn the wire made somewhere under the coil.
But as I noted in your other thread -- doesn't seem like much point in chasing the gauge behavior when you have coolant issues. If you are losing coolant I don't know how you expect you can make the gauge act normal. |
If you replace your controller with a newer short one (191 919 376A was what ETKA 7 update 597 specified for all years but there are at least two p/n that are functionally the same) you'll get low-coolant notifications without the gauge pegging - they send pulses long enough to trigger the blinker and often enough so it doesn't time out but not enough to move the needle noticeably, rather than the continuous ?30? ohms to ground that yours provides.
Highly recommended. |
|
Back to top |
|
|
LandSailor Samba Member
Joined: October 30, 2009 Posts: 315 Location: Las Vegas, NV
|
Posted: Mon Jul 28, 2014 6:57 am Post subject: Re: Help with blinking, Hoping to Leave on Tuesday |
|
|
dbeierl wrote: |
So you have a 1.9l. Tall or short controller? |
It's the tall one.
It used to occasionally "blink & peg" and I could reset it by killing the ignition. Now, it's doing it 100% of the time and will not reset. Pull the relay and there is no response from either needle or LED.
The 191-919-376A relay will flash the LED, but not peg the needle and require a power cycle, correct?
Also, so this condition does not rule out a bad cap? I wouldn't be surprised if it was the cap since (surprise!) the clock also doesn't work.
Just to clarify, when checking for the +10, you're referring to the electrical connection provided via the nuts on the flex PCB, correct?
Man, a bad voltage regulator sound pretty effin' sweet right about now! |
|
Back to top |
|
|
crazyvwvanman Samba Member
Joined: January 28, 2008 Posts: 9918 Location: Orbiting San Diego
|
Posted: Mon Jul 28, 2014 7:26 am Post subject: Re: Help with blinking, Hoping to Leave on Tuesday |
|
|
Does this mean you pulled the relay while the led was blinking and the needle was pegged and both things continued to be that way with the relay out?
Mark
LandSailor wrote: |
It used to occasionally "blink & peg" and I could reset it by killing the ignition. Now, it's doing it 100% of the time and will not reset. Pull the relay and there is no response from either needle or LED.
! |
|
|
Back to top |
|
|
LandSailor Samba Member
Joined: October 30, 2009 Posts: 315 Location: Las Vegas, NV
|
Posted: Mon Jul 28, 2014 7:35 am Post subject: |
|
|
Pulled relay, turned key on, dead LED/needle. |
|
Back to top |
|
|
dbeierl Samba Member
Joined: April 22, 2009 Posts: 164 Location: Providence, Rhode Island, USA
|
Posted: Mon Jul 28, 2014 7:37 am Post subject: Re: Help with blinking, Hoping to Leave on Tuesday |
|
|
LandSailor wrote: |
dbeierl wrote: |
So you have a 1.9l. Tall or short controller? |
It's the tall one.
It used to occasionally "blink & peg" and I could reset it by killing the ignition. Now, it's doing it 100% of the time and will not reset. Pull the relay and there is no response from either needle or LED. |
Does "no response" mean it doesn't blink/peg or it doesn't stop blink/pegging?
If it's doing blink-and-peg that's caused by the controller or short circuit somewhere on the sender line. The blinker circuit in the gauge is very simple: at power on and any time the resistance to ground at the sender terminal is </= about 35 ohms even momentarily, it will blink. Once it begins blinking it will continue for 2-3 seconds regardless, so if it gets a blip every second it will never stop. The timeout is regulated by a 10 uF cap inside the gauge, and when that gets leaky the blinking doesn't time out as it should. 35 ohms is the calibration point for the edge of the hot zone on the gauge.
Quote: |
The 191-919-376A relay will flash the LED, but not peg the needle and require a power cycle, correct? |
Yes.
Quote: |
Also, so this condition does not rule out a bad cap? I wouldn't be surprised if it was the cap since (surprise!) the clock also doesn't work. |
Argh. See above re no response. If the gauge is anywhere in the overheat zone then the light *should* be blinking. That's its only purpose in life: to blink when the gauge gets a sender input corresponding to overheating or when powering up.
Quote: |
Just to clarify, when checking for the +10, you're referring to the electrical connection provided via the nuts on the flex PCB, correct? |
That's right. That's the gauge itself so getting +10V there is absolute proof that the gauge voltage is ok. Not getting +10 could be regulator or various contact problems at gauge, regulator, panel ground, panel power. |
|
Back to top |
|
|
crazyvwvanman Samba Member
Joined: January 28, 2008 Posts: 9918 Location: Orbiting San Diego
|
Posted: Mon Jul 28, 2014 8:17 am Post subject: |
|
|
The temp gauge led should blink a few times when you turn the key on, whether the coolant level relay is installed or not.
If you unplug the wire from the temp sender and measure the voltage on the wire that should be 10 volts coming from the dash voltage regulator.
The capacitor that goes bad in the clock has nothing to do with the temp gauge. The temp gauge led blinker circuit has its own capacitor. Both capacitors are known to fail but they only affect their own device.
Mark
LandSailor wrote: |
Pulled relay, turned key on, dead LED/needle. |
|
|
Back to top |
|
|
dbeierl Samba Member
Joined: April 22, 2009 Posts: 164 Location: Providence, Rhode Island, USA
|
Posted: Mon Jul 28, 2014 8:21 am Post subject: |
|
|
LandSailor wrote: |
Pulled relay, turned key on, dead LED/needle. |
OK! That's closer to a defined state, though according to what you say the gauge doesn't blink at power-up which frankly I doubt). Gauge sender isn't the problem. Sender wiring isn't the problem. Internal cap isn't the problem. What's left is either
a) Controller operating correctly but input not being sufficiently grounded through the level sender line (bad contact, dirty pins, bad ground, weak coolant), or
b) Bad controller.
If when you first turn on the ignition you get a few seconds of blink that stops, then a short time later the gauge pins and blinks, the controller is probably ok. Not 100% sure but I think very strong indication based on my understanding of the internals of the controller.
If when you first turn on the gauge immediately pins and blinks, I believe that is proof of a bad controller. By the nature of the controller circuit it shouldn't be able to instantly have that output; it has to time out internally first. |
|
Back to top |
|
|
LandSailor Samba Member
Joined: October 30, 2009 Posts: 315 Location: Las Vegas, NV
|
Posted: Mon Jul 28, 2014 8:32 am Post subject: Re: Help with blinking, Hoping to Leave on Tuesday |
|
|
dbeierl wrote: |
Does "no response" mean it doesn't blink/peg or it doesn't stop blink/pegging? |
With the relay pulled, there is no needle movement nor LED. It's condition #3.
Upon re-reading the original troubleshooting post, I realize where I have a crossed mental wire. Initially, I pulled a lower relay (#4?) that also had a yellow/red wire, so I thought it was Condition #1 and was following that path. It was only after I pulled the fresh air vent that I saw the second row. (It was late and there's nothing like exhaustion and the fate of the vacation hanging over your head to make you miss things.)
I have a new relay that will be here tomorrow at Meyer's Auto Parts. Woo!
I will break out the Bentley and start tracing it back from the cluster when I get home.
On a side note, this may seem like a cheap way to do it, but it's funny how often it's used. I used to service lasers. The Coherent Innova 300 series (about $75-$90K, IIRC) take 480VAC, rectify it to 600VDC, and cool both the laser tube and passbank transistors with chilled water. The AC filament transformer (a coil of tungsten like in a lightbulb) has to warm up the filament before it starts throwing all those electrons down the tube. The start delay (warm up time before the DC flows) is controlled by a 555 timer chip that fills up a capacitor to ground out a transistor and activate the main contactor.
Thanks again for the help.
Last edited by LandSailor on Mon Jul 28, 2014 8:35 am; edited 1 time in total |
|
Back to top |
|
|
LandSailor Samba Member
Joined: October 30, 2009 Posts: 315 Location: Las Vegas, NV
|
Posted: Mon Jul 28, 2014 8:34 am Post subject: |
|
|
dbeierl wrote: |
LandSailor wrote: |
Pulled relay, turned key on, dead LED/needle. |
OK! That's closer to a defined state, though according to what you say the gauge doesn't blink at power-up which frankly I doubt). |
Sorry. It does work during initial startup. I skipped mentioning that since I didn't see any mention of that ever happening. |
|
Back to top |
|
|
LandSailor Samba Member
Joined: October 30, 2009 Posts: 315 Location: Las Vegas, NV
|
Posted: Mon Jul 28, 2014 8:48 am Post subject: |
|
|
crazyvwvanman wrote: |
The capacitor that goes bad in the clock has nothing to do with the temp gauge. The temp gauge led blinker circuit has its own capacitor. Both capacitors are known to fail but they only affect their own device.
|
I was only referring to the fact that all of the capacitors experience the same environmental conditions. My former boss bought a bunch of Dell computers in a lot from eBay. They were great for about a year, then Capacitor Plague started knocking them out, one by one. |
|
Back to top |
|
|
dbeierl Samba Member
Joined: April 22, 2009 Posts: 164 Location: Providence, Rhode Island, USA
|
Posted: Mon Jul 28, 2014 9:16 am Post subject: Re: Help with blinking, Hoping to Leave on Tuesday |
|
|
[quote="LandSailor"]
dbeierl wrote: |
On a side note, this may seem like a cheap way to do it, but it's funny how often it's used. |
In my youth it would have been an extremely fancy way to do it; more likely a thermal delay relay would have done the job.
Quote: |
I used to service lasers. The Coherent Innova 300 series (about $75-$90K, IIRC) take 480VAC, rectify it to 600VDC, and cool both the laser tube and passbank transistors with chilled water. The AC filament transformer (a coil of tungsten like in a lightbulb) |
First time I've heard of a transformer wound with tungsten! Seriously though, are you talking about a coiled coil as used in house lighting? I'm used to filaments folded up and down in a usually coated tube that's the actual electron emitter, or a similar bundle or straight wire as direct emitter. Incidentally as a kid I once tried to measure a 3VAC CRT filament floating on 1800VAC acceleration potential, using a metal-cased VTVM with negative input direct to the case. It didn't end well. My downfall was the metal setscrew on the range knob plus not knowing the one-hand-in-pocket rule - I was fine until I went to switch ranges. I have no idea what the common-mode voltage limit of the meter was, I lucked out on that one.
Quote: |
before the DC flows) is controlled by a 555 timer chip that fills up a capacitor to ground out a transistor and activate the main contactor. |
Inside the gauge they use a uA741 as output driver feeding a 5V self-blinking LED in series with a ?5.2V? zener diode. IIRC they time it out using some mishmash of a couple transistors, a big cap and a few resistors, with another zener in there somewhere as a reference.
Quote: |
Thanks again for the help. |
You're welcome. |
|
Back to top |
|
|
LandSailor Samba Member
Joined: October 30, 2009 Posts: 315 Location: Las Vegas, NV
|
Posted: Mon Jul 28, 2014 10:42 am Post subject: |
|
|
The tungsten filament is a coil that's about an inch long and looks like it's wound around a pencil. The concept garnered many perplexed looks ("wait. It's AC, but it's providing B+???") until the Iranian engineer explained the "boil off electrons" bit. Don't get me started on missing one question on the test because I couldn't understand how the difference between pronounced "walk in procedure" and "rocking procedure". |
|
Back to top |
|
|
dbeierl Samba Member
Joined: April 22, 2009 Posts: 164 Location: Providence, Rhode Island, USA
|
Posted: Mon Jul 28, 2014 10:45 am Post subject: |
|
|
LandSailor wrote: |
The tungsten filament is a coil that's about an inch long and looks like it's wound around a pencil. The concept garnered many perplexed looks ("wait. It's AC, but it's providing B+???") until the Iranian engineer explained the "boil off electrons" bit. Don't get me started on missing one question on the test because I couldn't understand how the difference between pronounced "walk in procedure" and "rocking procedure". |
Excellent, thanks! |
|
Back to top |
|
|
Ahwahnee Samba Member
Joined: June 05, 2010 Posts: 9797 Location: Mt Lemmon, AZ
|
Posted: Mon Jul 28, 2014 2:14 pm Post subject: |
|
|
If the blinking light is your main concern -- when mine began a false blinking a 1000 miles from home I loosened the nut right behind the light that holds the foil against it. This disabled the light.
Loosening worked for me but you may have to remove the nut entirely depending on how the foil is there.
Even with the light disabled, the temp gauge still worked which was fine by me -- that is the important thing. I can & do check the gauges as I drive so a blinking light to get my attention is not all that critical.
Note -- this was on an 84 with the big analog clock, which I think is the instrument panel you also have. |
|
Back to top |
|
|
dbeierl Samba Member
Joined: April 22, 2009 Posts: 164 Location: Providence, Rhode Island, USA
|
Posted: Mon Jul 28, 2014 2:25 pm Post subject: |
|
|
Ahwahnee wrote: |
Even with the light disabled, the temp gauge still worked which was fine by me -- that is the important thing. I can & do check the gauges as I drive so a blinking light to get my attention is not all that critical |
But you do lose function because there's no unequivocal indication of low coolant in the pressure bottle, and without coolant the gauge will respond slowly to an actual overheat. Much less stress on the aluminum engine to not let it get that far. Granted that with the old controller once you get the indication you still have to open up the hatch and distinguish overheat from low coolant; but the newer controllers fix that. |
|
Back to top |
|
|
LandSailor Samba Member
Joined: October 30, 2009 Posts: 315 Location: Las Vegas, NV
|
Posted: Mon Jul 28, 2014 2:32 pm Post subject: |
|
|
Why isn't this replacement controller on the Must Do List for new Vanagon purchases? |
|
Back to top |
|
|
dbeierl Samba Member
Joined: April 22, 2009 Posts: 164 Location: Providence, Rhode Island, USA
|
Posted: Mon Jul 28, 2014 2:41 pm Post subject: |
|
|
LandSailor wrote: |
Why isn't this replacement controller on the Must Do List for new Vanagon purchases? |
<crickets>... |
|
Back to top |
|
|
|