rebuilding H-D Delco-Remy cutout relays

[awander]

In case anyone is interested, I have posted about my rebuild of some cutout relays on the hydra-glide.com site.

[parkerize]

cool thanks for posting andy

[Chris Haynes]

Care to post it here for those who don't visit there?

[T. Cotten]

Chris!

Awander's discussion has a lot of pics, and this site makes that a genuine PITA!

http://www.hydra-glide.com/phpBB3/vi...a4112ec566a6a3

I thought you used to be registered at Hydra-Glide.net.
After tech threads have run their course, they are compiled into the site's Knowledge Base for easy reference.

Ever try the search function on this forum?

....Cotten

[awander]

Hi Chris:

Cotten nailed it-because of ease of posting, and also because of all the help I got from folks "over there:, i want to keep the post on hydra-glide.com. i really don;t want to deal with multiple postings.

I'll send you a pdf if you can;t get on the forum over there.

[Chris Haynes]

It wouldn't be much of a problem to cut it from there and paste it here.

[awander]

I am going to break this up into disassembly, rebuilding/maintenance, reassembly, and setup/testing.

Part I-Disassembly.

There are 3 soldered connections that will need to be undone before you can take the relay apart-these are:
1) series(heavy) winding to terminal
2) signal light wire to relay base
3) shunt(fine) winding to relay base

I recommend that you obtain a large soldering iron for this use-a large iron, like the one pictured below, will ensure that the iron has enough heat to quickly melt the solder joint; with a smaller iron you will end up heating stuff up for a long time, and getting the parts really hot, just to melt the solder joints. A large iron will allow you to heat the solder joints and the immediate structure around them quickly.

MVC-161S.JPG (57.02 KiB) Viewed 143 times

Heat the solder joint, and use a small screwdriver to pry the associated wire up and out of the pool of solder, thereby freeing it.
#1-Series winding to terminal:

MVC-162S.JPG (50.47 KiB) Viewed 143 times
After freeing the wire, bend it up so it points straight through the hole in the relay base.
#2-Signal light wire to relay base:

MVC-163S.JPG (49.47 KiB) Viewed 143 times

#3-Shunt winding to relay base:

MVC-168S.JPG (56.78 KiB) Viewed 143 times


Once those 3 connections are free, remove the contact point/swinger, as shown below. First, remove the 2 screws through the rear of the contact point/swinger assembly, that secure it to the armature:

MVC-165S.JPG (49.35 KiB) Viewed 143 times

Next, remove the top plate, the contact point/swinger assembly, and the spacer. Note how the angled end on the spacer is used to clear the 4th solder joint, that of the series winding to the armature frame:

MVC-166S.JPG (58.71 KiB) Viewed 143 times

The next step is to use a nutdriver to remove the nut, lockwasher, and flat washer that secure the armature and frame to the relay base:

MVC-167S.JPG (47.3 KiB) Viewed 143 times
At this point, you should be able to free the armature/frame from the relay base. A tap with a small hammer or a screwdriver handle, on the threaded armature stud, may be necessary to free it. If you give it anything other than a light tap, put the nut back on the end of the stud to protect the threads.

MVC-169S.JPG (53.87 KiB) Viewed 143 times

Now for the 4th solder joint-the series(heavy) winding to the armature frame(you can see the joint at the top left of the frame in this photo-there is a groove in the armature frame that the wire rests in before being soldered on:

MVC-170S.JPG (46.21 KiB) Viewed 143 times


Now, we need to free the armature(coils and core) from the armature frame. I would prefer to press it off, but I haven't figured out a way to hold the frame while pressing the core out-so what I do is to hold the armature frame in the jaws of a vise(I tried a smooth-jawed machinist's vise, but it didn't hold, so I used my serrated mechanic's vise as shown in the photo), thread the nut onto the stud, and gently tap with a small hammer-the core/stud will easily move out of teh armature frame(I have done 3 so far, and it comes out pretty easily):

MVC-171S.JPG (54.33 KiB) Viewed 143 times

Next, unwind the series(heavy) winding from the armature(remove any tape that covers it first), by unwinding it clockwise(looking from the top of the winding). Count the turns as you unwind it. The relays I have disassembled use between 36 and 41 turns of enameled 16 Gauge magnet wire. Once the series winding is removed, you will see something similar to this:

MVC-171S.JPG (54.33 KiB) Viewed 143 times
This shows the protective wraps over the shunt(fine) winding, as well as the braided piece of "tinsel" that is used to connect the shunt winding to the relay base. Pull off the vertical piece of tape, and unwind the wrapped tape, until you see the soldered joint between shunt winding and tinsel.

Here is what you will see after pulling off the vertical protective tape, and starting to unwind the wrapped tape over the shunt winding:

MVC-177S.JPG (48.03 KiB) Viewed 142 times
carefully move the tinsel/winding joint out of the way, and complete the removal of the wrapped tape:

MVC-180S.JPG (51.52 KiB) Viewed 142 times

Next, unwrap the shunt winding, counting the turns as you go. This winding also unwraps CW, looking from the top of the core. I find that a small pill bottle helps here, as I can capture the free end of the winding between bottle and cap, and then wind it up on the bottle as I unwind it from the armature:

MVC-181S.JPG (54.04 KiB) Viewed 142 times

You will come to a point(after about 1200 wraps of 36 Gauge enameled magnet wire) where there is another protective wrap-this appears to be a very this cloth or paper-you may be able to salvage it as I did on this one,(but then I dropped it on the floor and accidentally stepped on it). There will be another complete layer or wrap of fine wire under this protective wrap-for a total of about 1300 turns.
From what I can tell, the "inner" end of the shunt(fine) winding is not soldered on in any way-it is just wrapped around the armature core, and gets it's electrical connection to the core(and thence to the armature frame, and the contact assembly) simply through mechanical means. I suspect, but cannot confirm, that at least part of this final wrap of fine wire was stripped of the enamel coating, in order to make the connection. 2 of the relays I recently disassembled had a steel core, and enough corrosion on the core that the shunt winding did not make electrical contact to the core.
When I go thorough the rebuild instructions(to follow) I will show how to install a small piece of wire to make this connection more electrically sound.

I hope there is at least some interest in this!

Cleanup:
I recommend NOT disassembling the contact points, terminals, and relay base any further, unless there is an absolute need. These pieces are riveted together, through various insulating spacers, and the potential for damage is too great. Of course, if the insulating spacers are damaged, and allow any of the terminals or contact points to short to the base or to each other, you will need to disassemble further, and replace the insulators, and then re-rivet the assemblies. You may be able to install small machine screws and nuts(with threadlocker) instead of the rivets, to make it a little easier, but rivets will give the most secure assembly.

Clean up any rust or paint that are on the parts, using a small wire wheel, scrapers, etc. There is no need to make the parts look perfect, but you don;t want to leave loose rust or corrosion hanging off the parts. An ultrasonic tank with non-flammable cleaning solution may also work. I have not yet tried it. I'd keep the armature core, with it;s cardboard end plates, out of any liquids.

Keep the wire wheel off of the actual contacts; you can clean these up, if deeply pitted, using a small points file. Take off as little material as possible. final burnishing with a contact burnisher will come after re-assembly.

[awander]

Inspection:
-Make sure that the cardboard endplates on the armature core are tightly attached. If they are loos, you may be able to glue them in place with some Super Glue. They need to be pretty secure during the winding of the new coils.
-Check to be sure that the brass "spring" on top of the armature core is firmly fixed in place. The core has serrations around the very top of the shaft(under the head) and these should engage the hole in the spring. If loose, you may be able to solder the spring to the core, so that it does not rotate freely.
-check to be sure there is no excess rust or other corrosion on the various parts, which could cause a direct short-circuit, or could come loose and cause problems later.
-make sure the cardboard endplates are as flat as possible. If heavily warped, they may cause a problem during the winding of the coils. If they are badly bent, soaking them in water, straightening them, and then allowing them to dry might work.

New windings:
I purchased some enameled magnet wire on ebay, in 36 and 16 gauge. It was pretty inexpensive, the 36 Gauge was 1600' for about $12.50, and the 16 Gauge cost about $25 for 128'(both prices include the shipping costs).

There are lots of ways to wind the coils. You can do it by hand, or using a variable-speed drill. I am a half-assed perfectionist, so I chose to do it using my homebrew CNC mill. I wanted to count the turns exactly, and I also wanted to coils to wind as smoothly as possible, so the CNC mill, with a rotary "B" axis, worked well. I have a program to figure out the code for this, if anyone is interested. Here is a picture of my quick-and-dirty setup, ready to start winding the shunt(fine) winding:

MVC-196S.JPG (53.41 KiB) Viewed 91 times

A couple of things to point out:
-the spool of wire is on a shaft, with a spring to provide some friction.
-there is a small v-pulley which moves back and forth to spool the wire evenly across the width of the coil
-the 36 Gauge wire is taped to the inside surface of the "top" cardboard endplate, then looped over to the outside of the endplate, and the excess taped down to avoid any tangles. I used cellophane tape
-Note the cellophane tape over the inside of the "bottom" endplate-I placed 3 strips of cellophane tape, in a triangle, to cover the hole and the slot in the endplate, to avoid having the wire snag on the endplate. I made sure the tape was securely pressed down on the chuck jaws.

The end of the fine winding that is taped in place will be wound directly on the armature core, and will end up being the "end" of this winding that attaches to the swinger/contact assembly. It is not easy to see in the picture, but this winding should be made with the armature core turning CW(as viewed from the top of the core). Here is the first wrap partially done:

MVC-197S.JPG (43.06 KiB) Viewed 91 times


I programmed this to pause after the first warp, so that I could install an insulating layer, as the original coils had. I used friction tape for this purpose, and since my tape was about 5/8" wide, I had to cut the width down to about 0.60" to fit inside the endplates:

MVC-198S.JPG (45.65 KiB) Viewed 91 times
Once the first wrap was completed, I wrapped it with the friction tape-just a little over one wrap in length:

MVC-199S.JPG (45.95 KiB) Viewed 91 times


Next, I restarted the machine, to complete the 1300 total turns of 36 Gauge wire:

MVC-200S.JPG (43.76 KiB) Viewed 91 times

MVC-201S.JPG (42.99 KiB) Viewed 91 times


When it was done, I added another wrap of friction tape, to hold the end of the winding in place:

MVC-202S.JPG (39.43 KiB) Viewed 91 times


Cut the wire from the supply spool, and remove the winding/core from the machine. At this point, it is necessary to add the wires from the ends of the fine wire coil, which will be used to connect it to the other parts of the relay. The original Delco-Remy coils used a tinsel or braided wire. I don't have any of that, so I used a 24 Gauge stranded, insulated wire. First, it is necessary to strip the insulation from the ends of the winding. The easiest way I have found to do this on fine wire, is to lay the wire on a smooth surface, and sand the insulation off using 220-grit or finer sandpaper. I flip the wire over once or twice while doing this. I left about 1/2-3/4"" of insulation where the wires came out of the winding, and stripped about another 1/2" past that:

MVC-203S.JPG (46.64 KiB) Viewed 91 times

Next, strip back about 1/2" or so of the insulation on the wire you will use to extend the coil, and wrap the stripped part of the fine coil wire around it. Solder this using a smaller iron:

MVC-204S.JPG (58.61 KiB) Viewed 91 timesTrim the wire you just soldered, and then press them onto the surface of the friction tape. Note:
-the wire coming from the inside(directly against the core) needs to be connected to the notch in the armature frame, along with the corresponding end of the series(heavy) winding. This wire appears at the top of the photo below. The wire from the outside end will go to the relay frame, diagonally across from the end mentioned above. Make sure to position them so that they are close to their final positions
-make a small loop in the fine wire, before pressing it into the friction tape. Then route the soldered portion vertically, so that the insulation starts at the bottom of the coil.
Finally, put on another wrap of friction tape to hold things in place:

MVC-205S.JPG (55.68 KiB) Viewed 91 times

In the photo above, you can see the small spacer I installed on the stud, to keep the bottom endplate in place, and to allow a larger diameter for chucking up while winding the cores.

Next, wind the series(heavy) winding. It is easiest to start this winding by poking about 2" of the 16 Gauge magnet wire through the hole in the bottom endplate, and then wind on about 40 turns of the heavy wire. This winding is also done by rotating the core CW as viewed from the top. When you are finsihed winding, put a final wrap or 2 of friction tape over the heavy winding to hold it in place:

MVC-206S.JPG (59.07 KiB) Viewed 88 times

Now, we need to press the core back onto the armature frame. I have done this 2 different ways:
-with a small arbor press and a socket that fitted over the stud
-as shown below, using a large acorn nut that fit loosely over the stud, and pressing in a vise:

MVC-207S.JPG (46.89 KiB) Viewed 88 times

[awander]

Note that any spacer which clears the stud can be used to press the frame back onto the stud.

It is now time to fire up the big soldering iron again. While it is heating, strip the insulation from the 16 Gauge winding, and from the lead wire from the fine winding, at the notch in the armature frame. I have found that the easiest way to strip the 16 Gauge wire is to scrape the insulation off using a scalpel or X-acto knife:

MVC-208S.JPG (52.91 KiB) Viewed 88 times

Then strip back the insulation on the lead wire, place it in the notch, and bend the 16 Gauge wire down over it:

MVC-210S.JPG (52.74 KiB) Viewed 88 times

Finally, solder both wires in place, trimming them on the back side of the frame.

You can now place the armature/frame assembly in place on the relay base, and re-install the flat washer. lock washer. and nut that hold them in place. Note that the remaining 16 Gauge winding end passes through the hole in the relay base. After tightening the nut, bend this winding end down to place it over the terminal, trim to length, strip the end, and solder it on:

MVC-212S.JPG (41.98 KiB) Viewed 88 times

Flip the relay back over, then trim the remaining lead wire from the fine winding to length, strip the end, and solder it to the relay base:

MVC-213S.JPG (45.61 KiB) Viewed 88 times

Re-install the contact/swinger assembly, making sure to install the angled spacer plate, then the contact assembly, followed by the top cover plate, and the 2 machine screws. Pass the wire from the signal light contact down through the hole in the relay base:

MVC-214S.JPG (56.11 KiB) Viewed 88 times

Finally, solder the wire from the signal light contact to the underside of the relay base:

MVC-215S.JPG (55.6 KiB) Viewed 88 times


If the underside of the base flange, where it bears against the engine case, is not clean and shiny, you can touch it up on a wire wheel.

To set up the relay, you will need a 0.020" feeler gauge. Refer to this photo:

MVC-216S.JPG (51.52 KiB) Viewed 88 times

Unfortunately, you can't see the "air gap" in that photo, but the air gap is the distance between the top of the armature core stud, and the metal contact plate above it. With the relay in the "off" position(no power) the air gap should be 0.020". If it is not, bend the upper contact point bracket(the one that lights the signal light) until you get 0.020" air gap.

Once that is done, check the main contact point gap(the gap between the points at the far right in the photo). Bend the Lower (Main) contact point bracket to achieve a 0.020" gap.

Finally, as a check, press the contact plate down, into contact with the armature core, and measure the resulting gap between the upper(Signal Light) contact points. It should be 0.020" as well, though this is not critical.

To adjust the pull-in voltage, I use a large variable DC power supply, capable of several amps of current. I hook it up as follows:
"Minus" to the relay base, "Plus" to the "GEN" terminal. I hook up a DC Voltmeter to these same spots, and slowly increase the DC voltage from the power supply, until the relay closes. Then I look at the meter. The spec calls for these relays is to close between 6.3 and 6.8 Volts. the closing voltage can be adjusted by bending the brass piece, under the spring. Bend it up(toward the spring) to increase the closing voltage, and down to decrease. The relay has just closed in the photo below:

MVC-217S.JPG (51.32 KiB) Viewed 88 times


That's it! I have now rebuilt 2 of these relays, and I rode about 60 miles today using one of the rebuilt units. I will want to put a few more miles on her before I say "This Works!", but all indications are positive.

I found a couple of things that I thought were strange, but didn't seem to affect the relay function in the end:
-on 2 relays that I disassembled, the shunt(fine) winding measured around 40 ohms. The wire measured .005" diameter, so it is 36 Gauge. I unwound and counted 1300 turns. Once I put on a new winding of 1300 turns of 36 Gauge magnet wire, the new winding measured about 62 ohms! i am guessing this can only be because of the friction tape that I used for an insulator over the first wrap. The friction tape, being thicker than the original linen or paper insulator, would make the following turns have a slightly larger diameter, and so would use more wire. I was initially afraid that the greater resistance, which means less current in the coil, and therefore less magnetic force developed, might cause a problem, but it seems to work just fine.
-on all 3 relays that I have so far disassembled, the inner end of the shunt(fine) winding had no apparent electrical connection to the armature core/contact plate. In other words, it was not soldered to the core, or to the armature frame, but was just wrapped around the core. The winding I disassembled were in poor shape, though, so this connection may have once existed, and for the relay to function, it needs to be there. It's possible that the original windings had some length that was stripped, and then wound onto the iron core, thereby making the connection. I chose to run a small piece of insulated stranded lead wire and solder it to the end of the winding, and to the armature frame.

Comments and criticisms are welcomed, as well as any questions on things I might have left out. I hope this will help somebody else who has a bad relay to repair it and get it working again.

Let me know if you want me to rebuild one of yours, as well.

[awander]

Hey Chris, don;t mean to pick on you, but it is funny how easy it is to say how easy something will be for somebody else to do.
I just spent 20 minutes fighting the forum software(limits on post length and frequency), and in the end, all the text is here, but none of the photos.

[koanes]

What a great effort Andy. You have certainly set the bar high in the sharing of knowledge category. I don't know how many people would actually spend the time required to resurect an original relay, but the idea that they can be rebuilt makes me feel better about the small stash I have hoarded for 30+ years. Thank You.

[awander]

Thanks,Kyle. i love old machines(that;s why I own my Pan), and the relay is just another old machine.

I really like the way they were designed to be rebuilt(both the bikes and the relays)!

[Glenn]

great post thanks for all the info

you have motivated me to send mine to you to rebuild while you're not doing anything

[awander]

while you're not doing anything

...when's that...?

[Glenn]

when ever is good for you. Now i'm only sending it to you so you could practice more it's all in the interest of you becoming more professional in your interest. It has nothing to do with rebuilding mine so it is perfect you do understand. I'm letting you do this to help you out