Bloo

Those have a European pattern, a very good choice in my opinion. You can have more range on low beam, while blinding oncoming traffic far less. The "E4" tells you they have European certification. Autopal lights are made in India, and the "4" tells you what country the European certification was originally obtained in (I don't know off the top of my head what "4" is, but not India). These take "H4" bulbs, which are available in many voltages and wattages. People who have not converted to 12V can also use these after installing appropriate bulbs.

Electronic ignition makes less than zero difference in the timing needed. One possibility is that the ring on the harmonic balancer has slipped, and TDC is not really TDC. That can be checked with a positive stop in CYL #1. With all plugs out, battery disconnected, you crank by hand with a wrench until the piston stops against the stop, make a mark, and then go the opposite direction until it stops and make another mark. TDC is halfway between the two marks. Another good possibility is that the advance mechanism in the distributor, mechanical or vacuum or both, is not working as designed.

The extra 0.77v is lost in the wiring and grounds. Of course we would always like less, but I don't think that is too far out of line. The voltage is way too high as I'm sure you noticed. Make sure the generator, regulator , etc are all grounded good, and assuming the regulator is mounted on the body, that the body is well grounded too. The regulator might be adjustable. One of those relays controls voltage. If one is wound in *only* small wire that's probably it. It is basically a relay across the 10.64 volts you saw, and that should have pulled it down and shut down the field, causing the voltage to drop, causing the relay to contact again, causing a charge, etc. It happens fast normally and the points buzz when they regulate. So, it either... 1) has a burned out or disconnected voltage regulator winding, and the relay isn't trying to pull down at all (least likely) 2) is out of adjustment (maybe) 3) is not seeing the whole 10.64 volts due to a difference in ground, in other words a not solid connection between the base of the regulator and the generator case. (If I were a betting man, I'd go with this, but its always a crapshoot). If you try to adjust it, know 2 things to avoid chasing your tail. First, the temperature matters. Check the shop manual. The voltage at the battery on a 6 volt car is usually about 7.4-7.6V (tenths do matter) with the regulator at room temp. The lower end of that range tends to be for cars with over-engineered charging systems (like post-1939 two brush systems of 30 amp or more) and the high end for tiny 15 amp third-brush-plus-a-regulator systems that just barely work. The manual will know what the voltage should be at what temperature. Secondly, the setting changes with the cover off. You have to have the cover on to test because the voltage will be different. The cover affects the magnetic field in the relays. It also might help to leave the lights on for a couple of minutes with the engine off before each try at measuring the voltage, so you don't get fooled by surface charge on the battery. The spring tension on the little spring on the voltage regulator relay is the setting. A really tiny bend at the stationary spring anchor changes it.

Flickering on bumps is probably not a pencil beam, and if it is, the driver needs to beam down. It may indicate a European-inspired low beam pattern, and that should normally look downright dim to an oncoming driver. The picture below shows why. Some patterns about like this exist in the US nowdays. European spec lights like this are legal in several states, and some federally approved designs even approach this closely. US spec sealed beams on the other hand were *required* to shine up in the oncoming driver's eyes, allegedly to light up overhead signs . You can't make this up. Someone must have thought it made sense in 1939, and maybe it did then. I imagine this idea has crept into some of the current US federally approved lighting. Most of the latest models seem to shine in the oncoming driver's eyes just like sealed beams do, only brighter.

A dirty DC pulsing voltage comes from there. It is separate from the output post, and there is only voltage present at "R" when the alternator is spinning. It is good enough to turn on the field relay. This is how the system turns itself on (and off). The Riviera used ignition power to turn on the field relay, instead of "R".

How was the rest of the shaping done?

If it isn't double flare, it is probably "threaded sleeve" and probably 1/8". That was the most common setup on American oil pressure gauges up unitl the mid 50s at least. probably longer. Look down inside of it, If there is not an obvious inverted seat at the bottom of the threads for a reverse flare, and instead it has either a taper or a radius down there, it is probably threaded sleeve. https://blackhawksupply.com/collections/plumbing-brass-fittings-double-compression If that's what it is, and if you are in North America, you'll probably have to use copper tubing (not that there's anything wrong with that). 1/8" bundy tubing seems to be non-existent outside of the UK in 2023.

It's automatic based on the real world numbers the ECM (ECU in modern terms) gets from the sensors. @EmTee is on the right track here.

@Jack Bennett, the damage you show in the pictures cannot be fixed with lead any more than it could be fixed with plastic body filler. Lead is soft and breakable much like plastic filler. As @JAK mentioned everything needs to be shiny clean. It also must be structurally sound. Today some "metal finishers" with shops full of special tools eschew the use of lead because they can repair the metal to be good as new. In the old days though, a typical crash repair was not going to be smooth enough for paint, and lead was used. It was also used at car factories to hide seams in panels. The pictured damage needs metal replacement. You'll have to get it clean. If you can run an acetylene torch, which is also what they typically used for lead, sometimes brass/bronze can be used with less than perfect metal. I've seen this in auto body textbooks from 70-ish years ago, but it is frowned upon today because it is difficult to get paint and or fillers to stick to it long term. It is a potential idea to have in your arsenal though. Some of the areas in your picture are too far gone structurally for even brass, and I think you will need to make patches. Good luck. I'm glad you saved this truck.

Yes.

That Iron is very close to what would have been used in 1939. The one they probably used on the starter was intended for telegraph, knob-and-tube house wiring, and radio. It would have a smaller tip and a bit less wattage. I have 2 or 3 of those because both of my grandfathers were in trades that used them. On starters and any other huge electrical connections, I use one about like yours. The only downside is the akwardness of the giant tip. It sure does work extremely well for things like this. One caveat though, don't just leave it plugged in forever, just long enough to get hot plus about 3 minutes or so. Then solder and then unplug. Never leave it plugged in and unattended. An iron this big was meant for things like copper roof work, sheet metal, etc. and those things imply acid flux. As such, the thermal cutout in inside it is going to be set to a temperature too hot for rosin flux. It is impossible to keep the iron tinned if the flux is too hot and burning. An iron must be tinned with solder to work well, and this is one spot where you may need to break the rules for just an instant. You can use plumber's flux to get the iron tinned, in other words make it "take solder". I prefer the water based plumbers flux to the grease, less messy. Killed acid for radiators works too. In 1939 they would have used sal-ammoniac, but not everyone has that laying around today. Whatever you use (assuming the iron doesn't just tin easy with rosin), you need to get the corrossive crap off of the iron before you solder on something electrical. Wipe the hot tip on a dripping wet paper towel or rag. Immediately after cleaning off the now shiny tinned tip, feed it some rosin core solder to help keep it tinned. The #1 rule of soldering is "melt the solder with the work, not the iron". The solder alloys with the surface of the metal you are soldering. It won't work if the iron is the only thing hot enough to "tin". You ALWAYS feed the solder to the work, not the iron. Well almost always. Here is one more place you can break the rules, but only for an instant. You can feed a *tiny* bit of solder to the iron when you are heating the connection up. This is all about heat transfer. It gets the rosin flowing out onto the connection, and helps transfer heat. One more thing, Bloo's rule of soldering : "Get in, get done, and get out". In your case, if you are soldering directly to a wire braid, there is an extra challenge. The solder wants to wick up the braid and make it stiff. Obviously you don't want that. Dumping a whole bunch of heat into the connection QUICKLY before the surrounding area starts to get too hot is the secret. The thermal mass of that gigantic tip will help it happen, and it is the reason I use a soldering iron about like yours for this job. I'd have the iron tinned, god and hot, and then I would lay the flat side of the tip against the outside of that connection (not the braid) for maximum heat transfer. Get the connection good and hot fast, feed it solder at the back side (side facing left in your picture, furthest away from the flexible part of the wire), and as soon as you see it "Flow out" on both the lug and the brush wire at the back of the connection, pull the iron away.

I said that because you said: Putting 12v to "F" terminal on alternator (not the wire) makes the alternator charge as much as it possibly can. If you do this with the engine running you should see the voltage at the battery rise higher than normal. Do not leave it this way for very long. This is a common test called "full fielding" to see if an alternator is capable of charging. Since yours was like this: we know your alternator can charge, and you probably do not need to do the "full fielding" test. Pin F on regulator is an output, pin F on alternator is an entry. The large post on the alternator is the output, and is connected always to the battery through a large wire. Pin 3 on the regulator, marked Rouge/Battery in the diagram is an entry. A small wire from the battery (rouge) connects to it. This is used to power the regulator and alternator field when the system is on. Pin 3 is also used by the regulator to measure the battery voltage. Pin 2 on regulator is an entry. Voltage applied to this terminal turns the field relay on. The field relay turns the regulator and alternator field coil on.

If you connect F wire to battery, alternator will charge as hard as it can with no regulation. Voltage at battery should go quite high. With regulator connected normally, regulator should automatically cut voltage back voltage on F wire so that it maintains approximately 14.7V at battery. Regulator must always be grounded.

Some tractors have it as original equipment. In a car, it's a hack. I've been there and I don't recommend it. If you don't turn up the voltage regulator, you'll have to keep recharging the battery and the battery will get sulfated and fail right away. If you do turn the regulator up, the battery will be fine, but your charging voltage in the car will need to be about 9.9 instead of about 7.4 and you will go through a lot of light bulbs. Heater motors don't like it either, nor the radio. If you find one of these 8 volt batteries in a car, and you put a new 6 volt battery in, always check the charging voltage to avoid trashing the new battery. Adjust the regulator if necessary.

Yeah, flames are straight out. The old rule is ONLY ROSIN FLUX on electrical things, no acid, or acid core solder, or plumbing flux, etc. Only ROSIN CORE SOLDER, and if extra flux is used, then rosin also. That rule may be changed a little because there are electronic specific fluxes today that are not technically rosin, and one of those would be OK. Rosin and similar fluxes appropriate for electronics won't work with a flame, They just burn, and the resulting ash fouls everything and makes the solder not flow. Lead/tin solder works best. 63/37 or 60/40 flow easiest. 50/50 and 40/60 are more viscous, and maybe a little more appropriate on something big like this, but also a little harder to work with. ANY of the above will do this job fine. Leadfree solder is a pain in the a**. It could do this job fine too, but if you are not comfortable soldering I wouldn't recommend it.

The armature is the part that spins, those are field coils that the brushes are soldered to. It is going to take a pretty big soldering iron or gun, but on the bright side, it is a better connection than something bolted or screwed.

The battery voltage when shutting down suggests it isn't overcharging, but it is very hard to say. I am going to assume this is positive ground because that regulator says 6V POS on it, and I can't remember what year Cadillac switched to negative ground. You might need to do a really good comparison of the wiring with whatever is in the shop manual. I'll bet that's not the type of regulator Cadillac put on it, but if it is 6V, positive ground, and 2 brushes there is no reason it shouldn't work. 1935 is real early for 2 brush. I know there were some 1936 LaSalles with a 2-brush system that was shared with Pontiac police cars. The regulator had 6 terminals. That would have been a negative ground system though, so probably not the same parts as a Cadillac. With a battery that is not low or run down, and the engine running faster than an idle, so that it has enough speed to really charge, check the voltage at the battery. Also check the voltage between the BAT terminal on the regulator and the generator case. The voltage regulator should be regulating, and holding the voltage down to about 7.4 or 7.6 at room temperature. The service manual should tell you what voltage the regulator is set at. If it is running away, going up to 8 volts or something, then the voltage regulator is not regulating for some reason. The lid needs to be on because without it the setting changes, and the regulator base always needs to be grounded. There is also a current regulator, but that should not be doing anything under normal circumstances with a full battery. It is probably the left relay in the second pic. It is probably set too high, because 1935 era generators are smaller than typical. That doesn't mean more charge current. Not really. The current regulator's only purpose is to limit the generators maximum output to a tiny bit over the generators rating. How much current actually gets drawn only depends on ignition, lights, accessories, and how dead the battery is. A peculiarity of generators is that unlike alternators they will happily try to put out way more current that they are rated for if asked, and will burn up trying. Normally the only time you see a current regulator working is when the battery is half dead and wants a whole bunch of current to recharge. The battery may ask for more current than the generator can safely give. The current regulator is there to protect the generator in that situation. But yeah, check the charging voltage. That will be the biggest clue.

I think I have seen those pictures before, and I think it is a 1914. Maybe I'm mistaken, but if it is the same car it is curious if the seller knew what it is then and does not now. In any event, it appears to be a model 32, also occasionally referred to as a model H. Hupp wasn't really holding to exact model years at that point, nevertheless it is definitely HCCA eligible. All model 32s are pre-1916.

In the voltage regulator assembly, the relay at the left is a "field relay" and the relay at the right is the "voltage regulator". There is apparently a mistake in the diagram of the voltage regulator. If you could see the rest of the alternator diagram you could see that one end of the field coil is grounded, and so the alternator requires voltage on the field terminal to magnetize the field and make the system charge. The contact arrangement on the regulator relay cannot be as shown in the diagram. It would not charge. In the regulator assembly, field pin "F" is connected to power through a resistor that limits current to the field coil, and results in a low charge (2 amps or so) normally. The upper contact on the regulator (shown grounded) is really connected to 12v somewhere in the regulator. When battery is low, or if the car is not started yet, there is insufficient voltage to pull the relay down, and so the field, which is connected to the center contact, is connected directly to 12V on the upper contact, bypassing the resistor, and the alternator tries to charge as much as it possibly can. This is normal when the battery is low. When the battery becomes charged enough that normal system voltage can be maintained, the points open, and the resistor I mentioned earlier is back in the field circuit, no longer bypassed. It is cutting charge way back, to probably only about 2 amps. But then, the system voltage instantly drops, the points re-close, and the system goes back into full charge mode. This repeats at a high rate. The contacts buzz, giving an effective charge rate somewhere between high and low. This is normal operation with a full battery, and how the system maintains it's normal regulated voltage. If you drive long enough like that with no lights or accessories on the battery will get completely full and the relay pulls down a little more, hanging out between the contacts with the system in low charge mode. If you continue to drive, the battery will get a slight overcharge just from the 2 amps, and the relay will pull down even further, contacting the lower contact which is grounded. This grounds the field so the alternator cannot charge at all. Instantly the system voltage falls, and the process repeats. The center contact buzzes against the grounded lower contact just like it did against against the upper one and the effective charge rate is somewhere between the 2 amps and nothing. Now, back to your question about why the wires are the way they are, it has to do with the field relay, and the fact that one car has an idiot light and the other does not, as @EmTee already mentioned. A field relay is an attempt to make the alternator system "turn itself on" as most generator systems did. Alternators have diodes and do not have or need a "cutout" to disconnect from the battery, but then a new problem appears. The field and regulator would be on all the time and drain the battery when the engine is not running. GM and Ford used a field relay to solve this. Chrysler used the ignition switch. Now comes the fun part . On a relay it is easier to hold a relay in the "on" position than to pull it to the "on" position. The field relay, as I mentioned earlier, is there to turn the regulator and field on. In a car with an idiot light (no ammeter), the light has one contact connected to 12V ignition power, and the other contact connected to pin 4 of the regulator. When the ignition is turned on, some current flows through the bulb, and finds its way to ground anyway it can. It glows. One of those paths is through the field coil in the alternator. This makes a tiny amount of magnetism in the field coil, and it is enough to make the alternator try to charge a little when it starts turning. The "R" terminal on the alternator is connected to the center of one of the three diode pairs in the alternator. This gives a dirty pulsing DC voltage. It is connected to pin 2. When the voltage gets high enough, it pulls the field relay down. It will stay there, because it is much easier to hold a relay down than to pull it. Now the battery is connected to the regulator and field. Normal operation begins. Incidentally, now that the field relay is closed, and the system is charging, pin 4, which is connected to the idiot light, is hot. The other contact of the bulb is still connected to ignition power. Since both contacts of the bulb are hot now, there is no difference between the two, and the light goes out. In the second diagram, on the right, they just turned the field relay (pin 2) on with ignition power and let pin 4 hang open. Nothing wrong with that. The condenser is for radio noise. I don't know which terminal it connects to.

If the holes in the axles are OK, check the fit of the stationary part of the pins (important), if that checks out, the rest should have been done with a reamer back then. After pressing the new bushings in the spindles, you can probably(?) take the spindles to a machine shop and get the bushings honed on a Sunnen hone to match the size of your pins. Or you could just have the machine shop press them in first. I have heard you can get a better job this way, using the tools with which a wristpin is fit, though of course it is traditional to just use a reamer so that will work too. A reamer for kingpins is special, and has a long guide pin to keep the bores straight and concentric. In theory, the axle would stay in the car, and only the spindles should need special tools and professional help. You would have to assemble it in the car afterward with the bearings and shims to make a proper vertical fit. By the way, the pin to bushing fit probably needs to be considerably tighter than the machinist will think. You might even have to argue about it. It should be a little bit draggy and need to break in. Any slop is magnified greatly by the time you get out to the wheel. The difference between new and worn out is almost nothing. All the above ASSUMES that the axle is 100% traditional, and doesn't have floating bushings or any such nonsense. I am not sure how Buick did that in 1932. Measure everything carefully to be sure you don't have a floating bushing kit, or an oversize kit, and that the axle holes are OK and have not been reamed for oversize pins, etc. Tighter is better.

Those springs just come off, don't they? I believe they're separate pieces. If you unhook one from it's little brush box so it can unwind, it will just fall out I think. To put one in, reverse the process. Wind it up about like the others and hook it in the brush box. There should be nothing to rivet for just a bad spring. If a piece of metal is broken off where the tail of the spring is supposed to hook, then I guess you would need to rivet(?), if you could get the spring box thing(?), maybe? But in that case, I think you would normally replace the whole plate with those spring box things already riveted on, and it would probably come with springs, but if it didn't come with springs you could just hook yours in there. I don't know either. Get the engine numbers and pics showing details together and ask. I am not as intimately familiar with Chrysler flatheads as @Rusty_OToole, @c49er, @keithb7 and several others around here, but I imagine there's help. The p-15/d-24 forums also have a lot of people who know Chrysler flatheads inside out. Once you have the info together it wouldn't hurt to ask over there too.

Don't do that. Starters don't work like that. You could easily spent the rest of your life trying to find something to even fit, and then you wouldn't know whether the gear mesh is right and it would probably trash your flywheel, and then you would have to take the engine out. In any event, something from an unrelated vehicle like a Massey Ferguson won't fit. That way lies madness. Just don't. Fix that starter. Is one of the tabs that holds the spring broke off or is it just the spring? If it's just a spring, buy one. Those springs are going to fit thousands of starters from the same starter manufacturer. If the tab is broke, buy a new end plate. That end plate probably fits a whole bunch of starters from the same starter manufacturer, too. See the circular thing in the center? It's an Oilite bushing. It will probably already be in the new end plate, but it is a separately obtainable part, so if it isn't there, buy one of those too. You need an "auto electric" or "auto electric rebuild" shop. Most areas have one. It's a local guy who rebuilds starters, alternators, etc. They sell parts, and it's usually really affordable. There could be exceptions, but usually expensive parts are for semi-exotic things, and that isn't. That end plate is going to be a dirt-common part. You DO need to be able to identify what starter it is for. I believe you said the engine is not original. Is there a tag on the starter with a manufacturer and a model number for the starter? That will do. In fact it is what you want, and is way more important than make, model and year. If not, you need to figure out what make/model/year the bellhousing and flywheel are. I would start by identifying the engine, and then try to figure out whether the transmission was changed too. It seems likely bellhousing and transmission would be original. Best if there's a tag on the starter. That's easier Throw all the parts in a box and go there and say you want to buy an end plate. They can figure it out from the numbers on the tag, but take the rest along anyway instead of just writing the number down because in my experience they always ask to see it. Get brushes I guess. If none are shorter than the one in the pic, your old ones would probably still work if put back in exactly the same positions. If any are shorter though, probably not. In some other thread of yours, the commutator looked like it needed turning (machining), but I don't remember for sure. The auto electric guy can do that for you if it's necessary, or you can do it yourself if you can find a lathe somewhere. If you put new brushes in, you should turn (machine) the commutator, and you might have to if the contact area on the commutator is not flat enough to get a good contact while the new brushes break in. You don't "undercut the mica" on starters like you do on generators, so turning (machining) is a fairly simple affair. The old brushes already fit the commutator even if it isn't perfectly flat, so there's that, but they might be too short to work. You'll have to look. There's gotta be spring tension on the brush, the springs can't be bottomed out on the brush holder.

The thing to pay attention to in these old scan tools is the software included. Software is in a plastic thing like an old video game cartridge that plugs into the scan tool. It goes in the bottom of an OTC. There are four of them in the lower left of @60FlatTop's picture. Early on there were separate ones for GM, Ford, Chrysler, Imports, etc. Later on, the "big three" American makers were all in the same cartridge. Every year a new one came out so MAKE SURE that one you buy has a GM (or a "big three") cartridge that goes new enough to cover your car. Yes, you had to buy a new cartridge or cartridges every year if you wanted to work on brand new cars. There is also the OTC monitor 2000. It is older and less desirable than the monitor 4000 in @60FlatTop's picture. The 4000 added some little capability, I forget what, but the main thing is the bigger display. With a scan tool on a computer controlled pre-OBD-II GM car you get live sensor data, not just codes. On a 2000 you can only see 2 things at a time. I doubt there's much difference in value today so get a 4000 if you get OTC, but on the other hand if you saw a 2000 at a near giveaway price, it does the trick in almost all cases. I never updated to a 4000 before OBD-II came along and changed everything, and this is what I did for a living. I figured I would get a Snap-On Scanner instead. It has an even bigger display than a 4000. Once in a great while only seeing 2 things at a time on my old 2000 was inconvenient, but usually not a big deal. **Having a new enough GM cartridge is the main thing**. I believe the 2000 and 4000 take the same series of cartridges. Snap-on and any other brands would take different ones. Also, make sure the GM ALDL connector is with the scan tool. It is a pigtail that unplugs, because there are different connectors for Ford, Chrysler, etc. If there's a GM cartridge with the tool the ALDL connector should be there, but it unplugs and people do lose things.... Hey @60FlatTop, is there any reason he can't use the paper clip trick to get the codes on a Reatta? @Jess W, don't fall into the old trap, once you get a code, into thinking that the code tells you what is wrong. It's almost true on newer models, but back in those days. The code just told you what to look at and test. Today cars have systems with many sensors that interact, and often just replacing whatever the code says happens to work. It usually didn't work back then. Those cars had only enough sensors to tell the computer what it needed to know to run the engine, and a code just meant that the computer got some reading that it should not ever see. For instance, if a code is something like "oxygen sensor stuck rich". it the sensor might be "stuck rich" because it failed, but it also might be that the engine is running extremely rich. Once you have codes, get out the manual, check the circuit, and see if the sensor is telling the computer the truth. Maybe the sensor is bad, maybe it isn't, but if it isn't bad you still have a clue where to look next.

The things to avoid are general red flags. Do not kid yourself that criminals are using their real name or place of business. Identities and personal information may belong to people who are not involved and have no idea. Personal info can be scraped from the web just as easily as pictures of parts or cars you do not own. Fake (or stolen) identities can be changed on a whim. Some common red flags: New account with few posts (red flag). You check the user's post history and most or all responses are to wanted ads claiming to have the needed parts (red flag with blinking red lights on it). All the parts being offered are for a wide variety of makes, models, and eras (red flag with blinking red lights on it and also on fire). Seller wants to take the conversation instantly off of the forum so the administrators can't see it, "email me at xxx@xxx.com" "Call me at xxx-xxx-xxxx" etc. (same as above, but with a loud marching band circling around the burning blinking red flag). To be fair, there are a few, very few, honorable legitimate sellers on this forum who do that last thing. I don't know why. I guess they just don't get it. In that case though, you should be able to figure out that they are legitimate sellers by googling old threads, and of course their contact info will rarely if ever change over a period of years. "You should contact my friend (insert random name here) at (random email address or phone number), he has what you need". This one CAN be legitimate, but please know that this is ALSO the hallmark of the most common scam. If someone in the forum is legitimately telling you of a common source for your make model and type of car, they probably told others in the past. Google will reveal old threads with the same forum users recommending the same known good sellers. Despite the possibility of legitimacy in this one, I have to say red flag, blinking red lights, fire, marching band, plus a huge spotlight shining around up in the air. Next one, excuses why current pictures of the parts (or car) cannot be taken, "I work on an offshore oil rig" has been used so much it has become a meme, but there are plenty of new excuses. For sure the parts or car will not be available for new photos, nor will they be available for inspection by someone local. Maybe they are already "boxed for shipping" . There is the corollary that you need to send the money to my (insert friend/cousin/ex-wife) in (some other state). To make this one a trifecta, other people are interested and you must send a security deposit (or the purchase price) RIGHT NOW via Western Union or some other money transfer method that offers no recourse in cases of fraud (red flag, blinking red lights, fire, marching band, spotlight, plus a really loud carnival barker with a megaphone).

It needs to be connected directly to the manifold. On cars with a pump. tee in before the pump. On cars without, the wiper port is fine.