Yeah, I just wrote those readings off as noise. I didn't think to measure AC. the relay is kind of a pain to get to with the charcoal canister in place, but next time I'm in there, I'll see if I can get some AC readings.

Well said.

That is why I love schematics. I've messed with that stalk switch about 20 years ago but no way to recall all that from memory. The 3 track rotary switch on the motor armature and the inertia brake function further complicate things. Schematics . . . Schematics . . . Schematics. Must have for troubleshooting.


FYI - I was able to locate all those schematics on the interwebz so I didn't end up having to pull my reference materials out of packing boxes. You're still a dick for making me find the schematics to help you . . . But then again, I do owe you for letting me drive your car!

Please do put up a proper "How To" write up once you get this whole thing sorted. It really should get pinned over in the classics section.

Here's the blurb from the 73' owner's manual on how intermittent works with the factory graphic

Click to view attachment

Didn't I post the full schematic of the circuit at the bottom of page 1? I do appreciate the help. It's good to have multiple eyes thinking through the problem. I may revisit a simple MOSFET circuit. I think it might be possible with just the FET, a couple resistors and a diode which would be simpler and fully solid state compared with the LVD module. Will start with LVD and go from there.

I still say we didn't need the full schematic to troubleshoot this problem though. We only needed to know how the S1 wire is supposed to behave because that is the only wire modified by connecting the pump. The switch diagram tells the whole story on that: Open in position "J", 12v, in Position "0", and grounded in all others. That provides enough info to work on solutions. The rest was was fun and interesting, and I learned a lot more about how to read these circuit flow style schematics, but not mission critical. Interesting through that the old style schematics don't provide nearly enough detail to even figure out that one wire. Doesn't even include the intermittent wire or position.

Click to view attachment

Thinking about bench testing the LVD module when it arrives. I think the critical tests will be that resistance between B+ and B-, and between B+ and Out+ need to be open with 0v on the brwn/blk wire. Resistance between B+ and Out+ should be near 0 ohms above ~10v on brwn/blk, and the resistance needs to drop without significant lag when the voltage is applied. Anything else I should be looking for?

I did . . . just because that is the way I'm wired (pun intended) and was trained.

Probably the biggest piece for me was understanding how you had the intermittent relay wired in. For me, I need to see the big picture 1st before I can understand the details of what you had done. Words aren't good at conveying electrical circuit information. That is why schematics were invented.

I'm just hoping that you come up with a nice solution that allows operation of both the electric washer pump AND intermittent wipers since that is where I want to end up.

If the goal is to have an open circuit at S on the intermittent relay, a very simple solution may also be just adding a diode inline. I tested a 5 amp diode in a simple circuit with a 12v bulb and the meter shows 1.3 M ohms to ground on the side that feeds the 12V, which is where the tap is presently the relay. It's not an open circuit, but may be high enough for the intermittent function to work in the J position. It'd be a quick and easy test. Just a thought.

Click to view attachment

Dan

Yeah, I think it's a difference between engineering and ecology. The global ecosystem wasn't built by people so there is not blueprint or schematic of how it works. The scale and seemingly infinite variables is beyond human comprehension. The global circulation model is the most complex computer model ever developed, and only explains a tiny piece of our biosphere. We are good at homing in on just the subsystems that could be affected by the thing we are trying to understand.



To be fair, a good solution already exists. The full Car Magic conversion kits is something like $70 and provides everything needed to have electric washers and intermittent working as the factory intended. It just requires more aggressive modifications, labor, and $ than what I hope to accomplish here. The Car Magic kit is sitting on my coffee table just in case...

If indeed there is little voltage on this net when the intermittent relay is active then the diode or two would be a simple solution.

You have me curious. Did a little reading to learn about forward voltage. Sounds promising. I should have an assortment of diodes in the bin. Any recommendations on what to try?

We recently used some 20A 50V Schottkys for isolating the turn signals flashing when using the remote key lock. Gross overkill, but they were $1.50 each, so why not add some insurance. Any idea how many amps the washer pump motor draws? As long as you have headroom for amps and the voltage, should be good to try.

Dan

GENIUS!

Dug through my spares and found an unopened package of 1N4001 power diodes. Good for 50v and [EDIT: misread the data sheet. Only rated for 1 amp so I need to source a beefier diode, but the proof of concept is valid]30amps so plenty of head room. Vf is pretty high at 1.1v which is good for this application. Hooked one in series on the brwn/black wire and fired up the washer pump. Ran perfect. Pulled the lever down to turn on the intermittents and waited. No movement but a groan was heard from the motor, stopped, and repeated again after about 10 seconds. Hmm, sounded horrible but sounded like progress. I hooked up a second diode in series and WALLA! All functions are GO!

Should I add a third diode for an extra margin of isolation? So stoked! So the conversion now is down to:

- ~ $20 for a washer pump.
- A couple feet of wire
- 2-3 diodes
- (1) 3-way connector
- (1) 1/4" female spade crimp connector with locking tab
- (2) 1/4" female spade crimp connectors
- (1) ring connector (forgot the size I used)

Maybe $25 all in?

@lesorubcheek in my short remedial reading on diodes, I read that Schottky diodes have exceptionally low forward voltage so probably not the best choice. It seems cheap power diodes with high forward voltage are a better option.

Thank you, thank you for suggesting this. So simple and elegant. I'll use the LVD I ordered on my snow plow so it will go to good use. That was part of the calculation in ordering it.

Sounds like the intermittent relay needs a couple of volts to engage. If 2 diodes work then fantastic. The 1N4001 may be ok to use as the motor may not draw that much current. Also as you get higher current diodes the diode needs more current to get the same voltage drop so a larger diode may not work well.

You are correct about Schottky diodes; they have much lower ON voltage and not well suited for your application.

@Spoke

Oh right. I guess the "groan" I heard with only one diode was relay chatter and as long as the relay gets enough voltage to latch, it will be fine. I was worried about under voltage on the wiper motor and had forgotten about the relay. I confirmed that 3 diodes still work but seems to not be necessary.

I measured 4.4 amps draw on the washer motor which was higher than I expected, but in line with what I could find for similar washer pumps online. I noticed the 1N4001s were warm after running intermittent for a few minutes, which is what prompted rechecking the specs. I'm thinking these BY880-50 diodes might be a good way to go. Same "ON" voltage but rated for 8A, or would 5A diodes like the BY550-100 be a better choice?

@bbrock

Either BY880 diodes will work. According to their datasheets, both drop between 0.8 to 1.0V with 4A current.

I didn't see a price for the -100 part. The -50 part has 15 piece minimum which is no big deal. Both are 'marketplace products' which means Digikey lists them but they come from another supplier/distributor. They usually require minimum quantities.

Getting close.

Thanks. Yeah, I noticed the marketplace and minimum qty too. The only digikey diode I could find that is in their stock and matched my specs is 1 kV and over $6 each. Mouser has the BY diodes in stock with no min but doesn't off USPS first class postage like Digikey. Not sure if I'll be able to get that through the market place anyway so might must order some additional supplies to so I don't pay $12 to ship a $3 order.

Thanks for your help!

@Superhawk996 I just picked up replacement tags for the car so you can stop looking for junk by the roadside next time you pass through. The new tags even end in '73' like the old ones.

Unbelievable! The diodes arrived yesterday. I wound up ordering these from Mouser because Digikey's system was screwed up and wouldn't let me add the other diodes to my cart.

My plan was to do a quick test to make sure they worked, solder them into the wire, and be done with this project. So I twisted the diodes into the wire and pulled back on the wiper lever and yep, the washer pump ran although I could see it was spraying weaker than it had been. Pulled the lever down and the intermittent wipers ran. Looked good. Moved the lever back to the off position and pulled back on the lever again. Nothing.

Pulled one of the diodes out to see if just one would work. Nope. No washer, but intermittents worked. Broke out the DMM to start trouble shooting. I was getting 11.9v at the pump end of the wire with lever pulled back. Pulled the pump out and hooked it up to my power supply set to 12v. Nothing. The POS URO washer pump that came with the Car Magic kit has crapped out after only a few uses I wish I could say I was surprised, but there are many reviews of these pumps online of similar experience. Of course, I had to pull the pump apart to dig deeper.

Click to view attachment

The motor runs when disconnected from the pump. It struggles but runs when the motor shaft is inserted through the rubber seal in the pump, and chokes when brass gear is on the shaft even if the other gear is missing. Couldn't get the motor out without breaking the housing so that's as far as I went. I'm just assuming the pump puts too much load on a cheap-ass motor. [Edit: the pump is fine, I'm just dumb. Read on for what really happened.]

Anyway, the testing with the DMM showed that the 2 diodes wired in series will work, so I went ahead and soldered up the pump power wire.

Click to view attachment

I have a VDO pump on order so will have to wait for that before I can finish this up.

Well that escalated quickly
The new VDO pump came yesterday so this morning I went out to the garage just like last Saturday, thinking I'd spend 15 minutes tops swapping out the windscreen pumps and declare victory on this project. Two hours later I had this:

Click to view attachment

Click to view attachment

So what happened? Well, when I installed the new pump, it wouldn't engage pulling back on the wiper lever. More curious, the flick wipe feature was being reluctant too. I pulled the pump and hooked it up to my power supply and couldn't get it to run. I knew it was unlikely both the URO and new pump were bad, so I did what anyone with half a brain would have done last week - hook it directly to the battery. Yep, fired right up. Grabbed the URO pump and did the same. Yep, works fine. Turns out my cheap power supply just wasn't pumping enough amps to get the pumps going. Could have saved $50 had I done that last weekend. However, the VDO pump is much quieter than the URO. The URO screams so loud, it made me jump a couple times when I used it while driving. So I'm not that upset I bought the VDO. Anyway, so what the hell?

I unhooked the pump and checked voltage on the end of the brwn/blk wire again. 12v when the lever was pulled. Hooked the pump back up. Less than 1v. Hmmm. I cut open the brwn/blk wire so I could bypass the diodes to rule them out. Same results. I cut the diodes out completely just to make sure they weren't draining power somehow. Nope, same thing. I had an idea of what might be happening but continued to troubleshoot everything in the fuel compartment before acting on my hunch. Ground was good. Everything was connected correctly and working fine as long as the pump wasn't connected. Okay, time to pull the wiper switch.

I swear I took a pic of the switch inside the wiper control assembly but apparently not. Anyway, it is easy to see and is part of the same mechanism that pushes the plunger on the valve for the pneumatic washer gizmo. When the switch is in zero position and the lever is pulled back, there should be zero resistant between the terminals that connect to the red/blk and brwn/blk wires on the harness. I was getting varying resistance values, but always resistance. Then on the harness side, I connected a very small gauge jumper between the same wires on the connector socket on the harness and turned on the key. Flick wipers ran. Now I connected the pump and tried again. The little jumper got hot as hell and I yanked it. So my hunch was correct. The switch in the wiper assembly isn't rated for the amperage needed to start those pumps. I figured I'd probably fried the switch but I have spares so wasn't worried. On close inspection, it made more sense. There was a spot of carbon on the contact of the switch. Apparently running the pump directly off that switch was causing it to spark and build up carbon until the resistance was too high to flow enough current. After cleaning off the contact with a little 1000 grit polishing paper and a spray with contact cleaner, I was consistently getting 0 ohms resistance with the lever pulled back.

The solution, of course, was to add a relay so I grabbed the relay that came with the Car Magic kit and set about designing the circuit. Where to get power and ground without running wires all over the place which is part of what I wanted to avoid? Everything I needed was right there in the intermittent relay socket which would make for a nice, compact install. Here's the fully functioning test circuit. Pretty, ain't it?

Click to view attachment

Let's fix up some connections to tidy things up. The relay coil still creates a path to ground, so I still needed the diodes to fully disconnect the brwn/blk wire for the intermittent function to work. Since the brwn/blk was now just switching on the relay, I went back to the 1N4001 diodes because they are smaller so easier to solder and shrink wrap. Then a couple pigtail connections to split power and ground off the intermittent relay. I also changed the wire from the relay to the pump to red/blk to reflect it is switched power. Here's the full spread of all the bits. This could easily be done up as a kit. I'm not doing it, but if anyone else wants to, please do.

Click to view attachment

After tidying things up with the new connectors, I gave everything another test to make sure all was working before putting the fuel compartment back together. Washers worked, flick wiper worked, intermittent wiper worked... and kept working. Intermittent wipers never turned off. WTF? That meant the brwn/blk wire wasn't grounded in positions 0-2 as it should be. As soon as I looked inside the car and saw the wiper switch assembly dangling on the harness, I knew what the deal was. You know that weird little copper tab on the wiper switch that gets lost and causes trouble? Yeah, that's where the brwn/blk wire connects to ground. Mystery solved. After putting the column switches back together, everything worked as it should. Only thing left was to button everything up. Here's what the final installation of the relay looks like. Very stealth and very factory looking.

Click to view attachment

Click to view attachment

I'll write up a how-to on this. Was it easier than installing the Car Magic kit? Of course not. I like the install better though. Less drilling/molesting the car, less wire strung around, and everything is color matched to the factory scheme. It will also be a little cheaper and a LOT easier than the Car Magic kit for the next person. R&D can be a bitch. Thanks to everyone for helping work through this project. It was really pretty fun.

Very cool - and everyone likes a graceful recovery! Good troubleshooting.

A very interesting discussion. I have been kicking this around in my head for years, but had other priorities. However, years ago I 3-D printed a switch mount to replace the water valve on the wiper switch assembly. This gave me the trigger for the pump. I believe the switch (a tiny momentary pushbutton) does not have the necessary current capacity, so an additional relay would be needed to actually power the pump. I also 3-D printed a disc to fit in the washer tank cap to allow the tank to vent; this allows us to use the regular 914/4 washer tank with just a simple and reversible modification. All I have left to do is figure out the wire lengths and route it. I could have 20 sets of these parts available.
http://www.914world.com/bbs2/index.php?sho...7916&st=29#

I just realized I never posted a link to the CarMagic kit I keep mentioning in this thread. I added it in the first post starting the thread have added it here too. That kit sounds just like what you are describing, including he little disc to vent the bottle. I know several people have installed these and really like them.

I decided not to use it after reading SirAndy's tech article on adding an electric windshield pump. There I learned there is already an electrical switch inside the wiper switch assembly that can be used to trigger the pump. Even better is that most 914s already have a wire running from that switch to fuel compartment to trigger a relay when the wiper lever is pulled back. Figuring out how to get that to work with the optional intermittent wiper feature was a little tricky as the last four pages show.

The CarMagic kit requires too much drilling for my taste. This method only requires a single hole drilled to screw the pump to the chassis. The installation also doesn't require touching anything inside the cabin of the car except to pull or clip the water hoses from the switch valve, which I was able to do from under the dash. The rest is all done up front in the fuel tank compartment.

I'm kind of psychotic about wiring and making modification be as stealth as possible, so I used a little different approach for venting the bottle. I ran a drill down the small hole that is already in bottle cap and then drilled two small holes from the side into the channel created to turn it into a vented cap. Here you can see the drill bit sticking out the original hole and can just see it through one of the side holes.

Click to view attachment

I also left the tube from the cap to tire attached to the bottle to pay homage to the fantastically quirky pneumatic system that I regret having to replace. It's really there for the stealth effect, but does have a practical purpose in creating a tether for the bottle cap so it doesn't fall down into the fuel compartment when refilling the bottle. If I connected it to the spare tire, that hose also has a schrader valve in it that can be used to refill the spare without having to dig out you luggage and lift the frunk board. I'm not using that feature but might if I take a long trip with a frunk full of luggage. When everything is back together, this setup is completely stealth and looks like an ordinary spare tire powered washer system. Here's what my now electrified washer system using an unmodified stock wiper switch looks like.

Click to view attachment

Bookmarking this thread.

I may be able to piece together an original washer system if you like, cheap!!

I'm way past that of course. Countless hours and $ wasted trying to refurbish the old system. First problem was hoses. You need washer hose that can withstand 42 psi and it turns out that is not easy to find. I finally tracked down some NOS hose. Next problem was the valve in the wiper switch assembly. I have three wiper switches and the valves were blocked and wouldn't function. Try soaking in vinegar and Lime Away without luck so I bought a repro CIP1 valve. It leaks. Not into the cabin, but through the system so the washer nozzles constantly ooze liquid until the bottle was empty. In hind sight, I probably spent 3x as much time an money trying to get the original system to work as I did figuring out this one.

Now that the system is worked out, this conversion should be possible for $40-$70 in parts depending on which pump you choose, and ~2-3 hours if you are already familiar with crimping wire connectors etc., add an hour if you are not.

I do miss the brilliant stupidity of the tire system though.