75 - 79, 80 Fast Idle Control Valve "FIV" Working on a replacement

[79 Eldorado]

Just a couple random thoughts here.  I wonder what current/power at 12V your device will consume?  This is not very critical as the
circuit is large enough to run 2 pumps. 

Bruce,
I'm trying to stay very close to the original draw. I calculated just under 1 Amp. My assembly will have some internal parts acting as the heat sink which will be a little different in mass so I may need to vary some but I intend to stay close to the original.

What matters is roughly the right timing.  I note a couple different parts numbers for the small and big block engines, though it probably matters little. 

I had one of the original actuator you sent measured professionally. I then found an actuator wax which matches the progression extremely closely. I am pretty happy with that part.

Going backwards the Cosworth used a simple air valve that got 12V to add air, then abruptly closed at 125F coolant.  Just a step,
maybe no worse that dropping off a carb linkage step.  Cad has an IDLE STOP SOLENOID in the throttle body, wired to add some
throttle air when the AC compressor is engaged.  Probably not enough, wonder if it could be used like the Cosworth?  Bruce Roe

I remember you described the Cosworth air valve operation. That seems like it would be pretty abrupt. I think that "Idle Stop Solenoid" would be pretty easy to set-up. It could be relay and a temperature switch used to switch the relay. I don't know if it would be enough but it's an interesting idea. I'm getting close to a drop in. I still need the heating elements and control. I have a couple of options I'm trying to follow. Sometimes getting samples has been an issue. I do think I found appropriate control means for each of them. Once I have some samples of the heating elements I still need to design the interface between the wax actuator and the element. My thought is a custom brass sleeve with possibly an extended surface like a heat sink because the space around the wax actuator is small and some of the items which need to be in direct contact may be too difficult to fit without making use of the adjacent cavity.

Any thoughts on that clip? I don't see any reason for it. Even if it was for high volume assembly for anyone working on their own vehicle (not on an assembly line) it seems like it has no value... only headache if it gets "latched" or breaks and falls in the intake.

Scott

[TJ Hopland]

These work just like many of the electric chokes?   You get system voltage anytime the engine is theoretically on and it stays at full voltage till you shut the engine off?    You mention the fuel pump circuit?   Does this work like more modern systems where it won't stay on if the engine isn't turning?

[bcroe]

These work just like many of the electric chokes?   You get system voltage anytime the engine is theoretically on and it stays at full voltage till you shut the engine off?    You mention the fuel pump circuit?   Does this work like more modern systems where it won't stay on if the engine isn't turning?

The 70s Cad cold idle air valve is on the same circuit with the
fuel pumps.  And shuts off when they do, including shortly
after the engine stops turning.  Off, it will take time to go all
the way back.  But valve power comes out of the ECU on a
different pin, taking advantage of that has allowed saving so
many ECUs with burned out pump power pins. 

As for the clip, this electrical type has no idea of how to deal
with unlubricated sliding surfaces.  Bruce Roe

[79 Eldorado]

...I note a couple different parts numbers for the
small and big block engines, though it probably matters little....  Bruce Roe

Bruce,
I was able to search the GM parts wiki page for the Bendix PN on the FIV I had from you. It goes to a scan of the GM paper catalog. The Bendix number, 1181290, is listed as reference under GM PN 1606776 and it lists '76-79 "All w/E.F.I." (All meaning all motors and models). The next line does list a different PN for 1980. That PN is 1617956 and it also says "All w/E.F.I." (Meaning all 1980 as opposed to the line for 76-'79).

While I was researching this searched for any available photos I could find online. I did notice that the lever actuated micro switch on the side did have a different identification number on some versus others. I saw some with "E63" and some with "E62". When I dug further I found one number was capable of handling higher amperage than the other. The strange thing seemed to be the magnitude of the difference unless the source I found was wrong. I took pretty good notes and the E63 was listed as 0.1A and the E62 was a 10.1A version. Maybe there's something in that detail other than the maximum amperage because these FIV seem to need 0.92 Amps (at maximum nominal resistance). Anyway if GM had Bendix change the micro switch it would have almost certainly driven a different FIV assembly PN in my experience. That could be the reason for seeing other PNs. I doubt that's the difference for 1980 though because GM likely would have simply obsoleted a PN if that switch changed (not list two in the catalog).

I've also seen some photos of "white" (normally discolored to yellowish) plungers and others with black plungers. Maybe the color difference was associated with certain motors?

….As for the clip, this electrical type has no idea of how to deal with unlubricated sliding surfaces.  Bruce Roe

I didn't catch why the clip would help that. Are you thinking maybe it helped center the "plunger" in the "derby"? I know when I assembled the parts once on the TB it was really easy to slip and "latch" that clip. The wax motor has good pushing power and the spring it relatively heavy for returning the plunger once everything cools down. I just have a hard time believing the clip could have helped.

Scott

[bcroe]

Bruce,
I was able to search the GM parts wiki page for the Bendix PN on the FIV I had from you. It goes to a scan of the GM paper catalog. The Bendix number, 1181290, is listed as reference under GM PN 1606776 and it lists '76-79 "All w/E.F.I." (All meaning all motors and models). The next line does list a different PN for 1980. That PN is 1617956 and it also says "All w/E.F.I." (Meaning all 1980 as opposed to the line for 76-'79).

While I was researching this searched for any available photos I could find online. I did notice that the lever actuated micro switch on the side did have a different identification number on some versus others. I saw some with "E63" and some with "E62". When I dug further I found one number was capable of handling higher amperage than the other. The strange thing seemed to be the magnitude of the difference unless the source I found was wrong. I took pretty good notes and the E63 was listed as 0.1A and the E62 was a 10.1A version. Maybe there's something in that detail other than the maximum amperage because these FIV seem to need 0.92 Amps (at maximum nominal resistance). Anyway if GM had Bendix change the micro switch it would have almost certainly driven a different FIV assembly PN in my experience. That could be the reason for seeing other PNs. I doubt that's the difference for 1980 though because GM likely would have simply obsoleted a PN if that switch changed (not list two in the catalog).

I didn't catch why the clip would help that. Are you thinking maybe it helped center the "plunger" in the "derby"? I know when I assembled the parts once on the TB it was really easy to slip and "latch" that clip. The wax motor has good pushing power and the spring it relatively heavy for returning the plunger once everything cools down. I just have a hard time believing the clip could have helped.  Scott 

Maybe the clip keeps it from squeaking?

A 0.1 A switch is not adequate for the current, sounds like a
computer signal type.  Maybe they figured it out and changed
it in production.  I believe (in my collection of units with a
broken switch) I observed a couple somewhat different heater
resistances, presumed for different timing. 

The Cad valve may be much smoother closing, but the Cosworth
keeps working. 

Bruce Roe

[79 Eldorado]

Bruce,
Regarding the clip I think they just wanted the spring to be more compact/shorter during assembly.

I couldn't believe when I saw 0.1A on the micro switch. I wonder if what I saw was a misprint and maybe it was 1.0 Amp. They look so similar I couldn't believe one would be capable of 100x the current... doesn't seem to make sense but it doesn't matter now.

That Vega FIV solution may have lasted longer than most of their cars. Although to be fair the first and only new car my parents ever bought was a yellow Vega wagon. That particular car lasted over 200K miles. It was using some oil by then but it wasn't a disaster either.

Scott

[bcroe]

Bruce,

I couldn't believe when I saw 0.1A on the micro switch. I wonder if what I saw was a misprint and maybe it was 1.0 Amp. They look so similar I couldn't believe one would be capable of 100x the current... doesn't seem to make sense but it doesn't matter now.  Scott 

The switch difference is probably just in the contact.  Back then we
used a lot of switches (imagine rows of 24 for a 24 bit machine) in
computers, but the current was so tiny, it could not clean the contact. 
They would oxidize over at 3V.  So the contacts were gold plated, but
with very limited current capacity.  Same for these in some aps. 
Bruce Roe

[79 Eldorado]

Just a quick update: I was able to order components for a couple of different heating element and control schemes. Lead time is about 3 weeks.

The next piece I need to work on is a thermal mass/ sleeve to go around the actuator body. As well as providing some thermal mass it's also needed to provide enough surface for the heating and sensing parts. I didn't want to spend much time on that until I had some of the other items in hand but I may start on at least a version for one of the options.

Scott

[79 Eldorado]

I ran a baseline test on my car today. To me it seems normal. The RPM seemed to settle down in the time roughly reported in past threads but that said the current never got as high as I thought it would by voltage and resistance. It makes me wonder if one resistor after the mechanical switch is bad?

RPM provided by my timing light's tachometer, temp was taken at the water neck using a thermocouple (tried to make certain it was in contact but not 100% certain), DC Amps (reported as mA to be more visible on the same axis as temp) using a Hall effect sensing loop zeroed in place before starting. I was working alone so I had everything set-up but the timer started about 5 seconds after the actual start.

The temp in the photo to look at is the one labelled "KTC"

Not a big fan of this version of Office... I like the old graphing tool better but anyway it gets the point across.

Scott

[bcroe]

I ran a baseline test on my car today. To me it seems normal. The RPM seemed to settle down in the time roughly reported in past threads but that said the current never got as high as I thought it would by voltage and resistance. It makes me wonder if one resistor after the mechanical switch is bad?

RPM provided by my timing light's tachometer, temp was taken at the water neck using a thermocouple (tried to make certain it was in contact but not 100% certain), DC Amps (reported as mA to be more visible on the same axis as temp) using a Hall effect sensing loop zeroed in place before starting. I was working alone so I had everything set-up but the timer started about 5 seconds after the actual start.

The temp in the photo to look at is the one labelled "KTC"
Scott.

Scott, is this a run of OEM parts, or the new components you ordered
in April?  It looks like what is supposed to happen.  The resistor current
I do not understand, I expect a higher start current and then a step
down as minimum idle rpm is approached. 

Measuring current, my experience is that Hall clamp on devices have
poor accuracy at these low currents, and tend to start drifting (zero
set) before long.  My preferred way is to insert a low value resistor and
measure the voltage drop across it (shunt).  Perhaps I could send you
an adapter harness to insert in the heater feed, with a 0.1 ohm resistor
in the path you will see 100 mv across it per amp of current flow?
nice work, Bruce Roe
not very accurate at these

[79 Eldorado]

Hi Bruce,
That run was with the OEM parts. My car has always acted "normal" but my car is the only reference I have to this FIV type system.

I was also expecting to see exactly what you described. The meter I used does have a low set point, 2A, but point taken on the drift. The instructions say they need to be zeroed each time before measurement and I read that they are sensitive to magnetic fields. I did notice when I turned everything off there was a stray reading. I'm almost certain it read 0.126 Amps; so anyway not zero as was the case before starting. Given the car acted as it did it would be hard to believe it would have acted as nicely with only half of the heating.

I don't have any OEM type connectors to assemble an adapter harness but I like your idea and I could set it up if I had an adapter like you described. If you are willing to put one together I would certainly try it.

One of my friends said he might stop down this week in case I get in a situation where I wish I had an extra hand. I was fortunate the multi-meter was stable sitting but I would have liked to have someone watching as the car was started with the wires and monitoring devices instrumented.

I have everything needed to assemble the prototype now except the brass sleeve. Those are close. The temperature controller is a little different than what I was expecting but I think it will work.

As always thank you for your suggestions an your thoughts,
Scott

[bcroe]

Scott,

A current shunt to aid your project may get there Thurs.  You
can plug it into your wiring, read out the current accurately
with a DVM on 2 access wires.  thanks, Bruce

[79 Eldorado]

Bruce,
Received and working perfectly! I only added a couple of connectors to the green and white wires so my DVM leads could plug in and stay put to avoid another reason to wish for extra hands

The place I got the thermal actuator from added 5mm to the lengths I told them I wanted to try. I guess better too long than too short but the pin they use is very hard steel. I broke a tap trying to thread it while making an adjustable version for trial. After breaking the tap I decided to get the closest brass rod I could find. I created the replacement pin by turning the rod down/cutting to length.  I just tested the set-up you sent me with the first prototype on the kitchen floor. The replacement is drawing 0.9 Amps when it's on. The controller samples something like every 20 seconds and allows heating if the temp drops below the "on" set temp.

I likely need to do one more redesign of the housing because everything is very tight. The assembly was very difficult. I finally decided I would leave the top off for the trial. I think I will increase the height slightly but I will keep the terminals at the same height.

I also came across the comment in the 11x17" Electrical trouble shooting manual that the FIV should take between about 1.5 and 5 minutes depending on the conditions. My earlier trial with original 40 year old parts now seems as if it took too long even though my perception was the system was working properly.

Scott
Edit: Added photo with Bruce's harness hook-up for measuring current draw.

[bcroe]

That is great work, hope I can continue to provide some assistance.  Are
you planning to continue using the switch?  It occurs to me there might
be an electronic way to reduce current after warm up.  Remember the
color TV pic tubes, used a degaussing coil at turn on.  It was fed through
some thermistors that heated up and changed resistance to turn it off,
no moving parts.    Bruce Roe

[79 Eldorado]

Bruce,
There are some PTC heaters and I was thinking about using one with a fixed resistor to get the desired output. When it hits the proper temp the resistance would be so high it would essentially have no heating. I even bought some parts to try but the problem is space including mounting the heating device in thermal communication with the actuator. If I could dictate exactly the PTC heater size/ resistance maybe but there is a lot to pack in a small shape.

This set-up uses a smart controller and a temp sensor. I was able to select the programmed off temp and restart temp. That gives me minimal connections but the space is still tight.

I also had an idea for a dual stage but it required more connections and 2 separate heating elements. I also needed to extend my heat sink to mount everything. After I put this first one together I doubt I have room for the dual stage.

With this first controller design there will be no switching arm like the original. The number of on-off cycles is reduced by the staggered off-on temps. The controller is rated for millions of cycles. Now that I know everything works inside I can determine the time between switches. To make that time greater, decrease the number of switches, I made certain the heat sink was about as large as it could be. That also has an impact on the heat-up time. The first time I powered it on I didn't start from equalized to measure but after it achieved the target temp it seemed like the time before the next cycle was sufficiently long. I will probably take the measurements tomorrow. Now that I know the parts seem to be pretty stable I may try it directly on my car but the bench test is good to see if things seem to be working as normal.

I know I probably need to spend at least a day reworking the size to fit things a little more comfortably. It probably needs to grow only slightly but I may change how I secure the terminals when I do it.

You always provide great helpful ideas

Scott

[79 Eldorado]

Bruce,
I tested the FIV you had indicated as "OK". As you already knew the resistors checked out fine and the switch is working. I think the issue with these is the thermal paste is likely providing the thermal communication between the wax actuator and the heating element and when enough of that leaks out the thermal link is somewhat broken so it takes a really long time to heat the wax actuator.

So in this example I got good movement from the wax actuator (probably nothing wrong with the actuator). The total travel was about 4.3mm. The current shift due to the switch position finally occurred at 10min 19 seconds. I almost gave-up on it but it finally happened and it was obvious. So I think everything with that FIV works with the exception of likely there is little thermal paste still present. I was only bench testing it but I had it set-up in the spare throttle body with a depth gauge on the center of the "cross" (bottom of the plunger).

One other thing I noticed was the current was a little on the low side after starting at what I thought was slightly high. It started out at 1.19 Amps but within 4 seconds it was down to 0.94 A at 24 seconds 0.82 A at 46 seconds 0.79 A. There was some small drop after that every couple of minutes down to 0.66 A at 6 minutes where is stayed constant until just before switching. After the switch it went from 0.66A to 0.37 Amps. I haven't graphed the data yet.

I also tested my prototype with this set-up. I did it with the brass pin I made as well as the steel pin which came with the actuator. I think something slipped during the test with my pin. My pin is rough and I should try it again. I switched to the steel supplied pin to measure the travel and that was very close to the "OK" part. The target temp was reached in 1min 39 seconds. It was at least 80F here with a high predicted to be a record breaking 88F. Considering the Troubleshooting manual listed between 1.5 and 5 minutes I thought 1min 39 wasn't too bad considering the ambient and no airflow around it.

Scott

[79 Eldorado]

I ran the test with my brass pin one more time. It repeated the steel pin test. My dial indicator must have slipped during the first test. See graphs attached. The next test will likely be on the car with RPM. Since every original I have has a different pin length this first brass pin is a bit of a guess and I think I kept it just slightly longer than I expect it should be.

Scott

[bcroe]

Have you somehow replaced the missing paste?  Bruce Roe

[79 Eldorado]

Have you somehow replaced the missing paste?  Bruce Roe

I haven't tried to replace it. I thought about it but the thermal paste I have is electrically conductive so I didn't want to risk trying it. Based on your question I bought some non electrically conductive paste which should work. When it arrives we can see what happens.

One problem with the original design is it seems the only thing stopping the paste from seeping out is a tight fit but it's obviously not that tight. Even NOS parts I've seen photos of show signs the white paste is coming out. One person even stated it was normal; while true still not ideal. My design very likely doesn't require any paste. I decided to use some to keep the thermal connection consistent but I added an o-ring on the gravity side. I even made certain the o-ring was compatible with alcohol containing fuel. My thought was worst case there should only be residual fumes from evaporation but I didn't want to take any chances.

Scott

[bcroe]

It is evident to me, in an effort to quickly get out a then best state
of the art performance, they slapped together a lot of somewhat
revolutionary new stuff and did not do enough testing.  With the new
digital designs in sight, no more effort went into fixing these.  Now in
hind sight we are slowly working out all the bugs. 

My ECU work has gotten more complex than a very simple repair,
because I also try to head off other potential future failures. 
Bruce Roe