Experimental rail pickup construction: wiring question

[Bob Francis]

Jason please keep this thread going it's great!

[Jason Rodgers]

Yes, sir!

[Jason Rodgers]

Tonight, I began work on some bases for the coil pairs. Again, I'm using some of the laminate material that I have. I measured and marked the material, cut it out on the bandsaw, and shaped with disc sander and a file. The sticks are pieces of Doug Fir that will become spacers between the bobbin and baseplate. They were glued on with CA.

[Jason Rodgers]

Figuring out how to attach the bobbins to the base plates was not easy. First, I simply had to figure out how to hold it all in place.

It was pretty hairy, drilling right up close to the coils. I certainly have learned how NOT to build these next time. But eventually I got some holes drilled and attached the base plate with these wee stainless screws. These are only so I can disassemble things to play with the magnets and such until I know that it all works. Then I'll flood the space between the bobbins and base with CA or epoxy.

I still need to join and extend the ground wires (I'll be doing the twist method), but for all intents and purposes, these puppies are ready to rock! Better get that guitar finished, because it's not like I can just load them up in any old stock axe for testing. In the meantime, I can try wiring them up to a jack and holding them over the strings of another guitar. I'm so curious to hear them!

[Hans Bezemer]

Jason, they're looking great. Eager to hear a soundsample!

[Mark Swanson]

I would just pot the entire assembly when you are done, and skip that glue step you have planned. The glue probably wouldn't stick to the already potted coils anyway, and a loose plate will cause feedback and microphonics. The time for potting anyway is the last step, you should have just done it once after they were all assembled.

[Jason Rodgers]

Thanks, Hans!

Yes, I've considered that option, too, Mark. I've been reading on the MEF about different potting approaches. Some folks do an initial coil pot, and then dunk the assembled pickup to fuse it all up (to avoid the vibration of base/cover that causes microphonics). Some folks do this with wax, and others use clear penetrating epoxy (the kind used to stabilize rotted wood). Rick Turner reported doing both: potting the coils with wax, and then sealing it all with an epoxy. If I were to use epoxy, I was trying to figure out how to encapsulate the base and the bottom flat of the bobbin, over the wires, and maybe a little ways up the coils. Maybe some sort of tray or tape dam. Turner has a pickup cover with no holes, sets the pickup face down, and fills up the cover. Dunking the whole thing again would ugly-up the Kapton tape - it makes it cloudy - which will be visible in the installed product.

I have some time to consider my options, as it will still be a little while until the guitar is ready for electronics installation.

[Jason Rodgers]

Ug, well, I have good news and bad news.

Yesterday, I installed the pickups on the guitar they were made for, and wired it all up. I used a control arrangement that I enjoyed on a Carvin DC-127 I had years ago. Volume, tone (500k audio), .022uF cap, 3-way toggle switch, and 2 mini switches for coil tap (taps select north coil on the neck pickup and south coil on the bridge). Just to make sure everything was working, I plugged it in today. The bridge responds to a tap on the rails, but the neck does not.

I pulled out the multimeter and started checking continuity and DCR. Everything connects, but the DCR is wrong. With the toggle selecting the bridge, I touched my leads to ground and hot at the volume pot. This read as expected, with 13.6k in humbucking and 6.8k tapped. Then, I did the same for the neck, but got a very low reading. I chalked it up to a cold solder joint, so I clipped the leads at the neck mini switch, stripped, tinned, and resoldered them. Alas, no change. This time, I went directly to the leads of the neck coils. I should be reading 4.8k ohms on each coil, but I only get 2.4k. I'm really hoping I don't have a short.

Here's something else that's odd: when the toggle is selected to bridge, the rails respond to taps; when the toggle is selected to neck, no response; when the toggle is at center, NEITHER neck nor bridge are on. I'm thinking there's something wrong with my wiring scheme and that everything is getting sent to ground when the neck is selected. Or can a short in the coils ground everything, too?

This is basically a Carvin diagram with lead colors changed (to keep things straight in my reading, I used Lollar's lead colors, which also happen to be the same as Seymour Duncan). Please take a look and tell me if you see anything glaringly obvious (because I barely understand this stuff).

[Greg Robinson]

That all looks good to me Jason, and you've got opposite coil cuts for the two pickups so middle position with both humbuckers cut should still be hum-cancelling.
It sounds like you've inadvertently wired the neck pickup in parallel, then when you connect it up to the mini switch, either position will short circuit it to ground. Double check your wiring at the pickups themselves, I'll bet that's where your problem is.

[Jason Rodgers]

Thanks for the schematic check, Greg. But I'm certain that it's wired as you see it. When you say I may have made a mistake at the pickup, do you mean I got my lead colors wrong from the get-go? What would be a solution?

[Greg Robinson]

No problem Jason.
Given that you're reading 2.4k on each coil rather than the expected 4.8k, I'm guessing that the two coils are in parallel. You may have wired it incorrectly from the get-go (easy to do), or it's possible that there's a short somewhere in the cable, it's easy to melt the insulation on the tiny little stranded wires while soldering.
Frustrating, but I'm sure you'll figure it out.

[Jason Rodgers]

Grr. Just dug around a little more and am still coming up with the same readings. Here's what I did...

First, I triple-checked my lead colors. As you can see in the pictures here, the north coil starts black and finishes white; the south coil starts green and finishes red. Those finish leads are on the outside, so I know it's right.

Then, I checked the DCR again. Black to white reads 2.42k, green to red reads 2.41k, and black to green (through both coils, as red and white finishes are tied) is 0.02k. WTF?!

Finally, I pulled the pickup apart so I could access the finish leads (soldered to the magnet wire) and tested again. Same numbers.

This is obviously pretty disappointing, as everything was going so well. Maybe potting was too hot? I dunno. What sorts of numbers show up when you have a short? Are they so perfectly half of the starting DCR? And do you still get connectivity (the "logic" setting on the meter that beeps)?

[David King]

Check to be sure you don't have a short between the windings and the armature. If you have shorts between your starts and the armatures on both coils you would see a very low DCR reading when measuring both coils in series from start to start.

[Greg Robinson]

Yep, it's as I suspected, somewhere your starts are connected together, either in the pickup itself, or in the lead - black and green, so the two coils are in parallel. If you were to bypass the switch and use either black or green as one lead, and red/white as the other, you'd get output (but not humbucking, it'd be out-of-phase). I'd recommend you put a new lead on the neck pickup, you're sure to find where the problem is if you do that.

[Jason Rodgers]

David, my steel cores were sprayed with Krylon clear, then Kapton taped after assembly. I don't know how my starts could short through that. The starts are taped to the core so subsequent wraps can't touch the solder joint. They are buried under the coil. I could probably mine them out, but at the risk of tearing overlaying wraps.

Greg, I'm going to trust that you know what you're talking about, but I still don't see how it's possible that my starts are connected. If the diagram is correct, then I've followed it exactly (twice now).

Here's what I'll do, though, to use both of you guys' suggestions: tomorrow, I'll clip all the leads on the neck pickup and test each coil completely independently. If it tests correctly, I'll solder it back into the system, paying careful attention to the schematic. If it doesn't test correctly, I'll start replacing leads.

[David King]

Jason,

Have you determined with a meter that there is infinite resistance between the coils and their armatures?

[Jason Rodgers]

Ok, back out in the shop tonight since I might not get a lot of time tomorrow, and the clock is ticking on the GAL Convention!

I just snipped the neck pickup leads off the controls and tested everything in isolation. Upon initial inspection, the coils read the original correct resistance! However, I also tested across the cores and may have found my short. When testing green to core ground (south coil), I get nothing. When I test black to core ground (north coil), I get 0.02k. Furthermore, when I test white to core ground, I get the whole DCR for that coil. I'm guessing this means that all of my swearing up and down that there was no way the start could short just got disproven.

Well, this was the first coil I wound, so I guess it wasn't as brilliant as I thought. Time to dig out that start, pull a couple wraps out, and try again.

[Jason Rodgers]

Wise man once said, "You gotta know when to hold em, know to fold em..." I got about an inch of start lead pulled out and it breaks, the end lost under 5000 wraps. So, it looks like I get to experience my first rewinding this journey, as well!

[Jason Rodgers]

Ok, new coil wound, reading 4.8k, and no shorts. If I would have known to test across the ground when I started, I probably would have found this short right away. On that first wind, I bunged up a bunch of wraps and had to unwind a small pile at my feet. It all went back on smoothly enough, but I'm sure something got kinked or scratched enough to allow that short. The new coil, overall, is tighter and more neatly wound, so it's all good in the end. And heck, I kinda like winding coils!

Thank you, Greg and David, for jumping in on the trouble-shoot! This was pretty frustrating, but a great learning experience, too.

[David King]

I have probably burned through 5lbs of wire this way. It's just part of the fun. You can spend 30 seconds cutting the old wire off and 10 minutes rewinding the coil or spend days pretending that the "impossible" didn't happen. It happens all the the time to everyone. You get better at it if you're winding coils all day everyday but if I wind one a month my odds of screwing up a coil are closer to 25%. Most of the time it's shorts within the coil between layers and those shorts are much harder to find and don't help the sound at all.