The Early Univox Phase III humbucker pickup is the holy grail pickup for Early Nirvana Tone
Kurt Cobain the luthier, knew what he liked. The Univox Phase III pickup was his favorite of that Bleach era.
He would use this pickup on his first Mustang, SG and featured on the Bleach Album. Why did he love this pickup so much. The answer is possibly the resonant peak. A tad bit higher than a normal PAF. Brilliant for sure!
• Below 1 kHz: dull, hollow, muffled.
• 1.0 to 1.5 kHz: very soft, smooth, and mellow (think Clapton’s Cream-era “woman tone”).
• 1.5 to 2.0 kHz: warm.
• 2.0 to 2.5 kHz: singing—a typical PAF humbucker’s resonant peak.
• 2.5 to 3.0 kHz: brilliant.
• 3.0 to 5.0 kHz: harsh, shrill, metallic—the resonant peak for Strat and Tele pickups.
• 5.0 to 8.0 kHz: sharp.
• 8.0 to 15.0 kHz: glass-hard, ice-picky.
• Above 15.0 kHz: neutral, colorless, lifeless.
https://www.premierguitar.com/mod-garage-exploring-outlaw-pickup-parameters
I was luckily able to play a couple Univox Phase III guitars. One was a sunburst and the other was a Natural finish. Shoutouts to one of my customers to let me borrow these guitars for a few weeks to analyze. The sunburst guitar sounded amazing. The lows were very clear and articulate and captured what felt like magic to my ears. My inverted chords through a big muff really shined. The natural finish phase 3 sounded great also just didn’t quite have the mojo. That mojo was measured and analyzed using a USB oscilloscope to compare the resonant peak of both pickups. Also noted there is no wax potting.
The resonant peak shown above is 2.99 kHz noticeably lower than the Univox Pickups I tested. I tested a total of 6 Univox pickups. 2 from a sunburst phase III, 2 from Natural phase III, One purchased for 200+ by customer, origin unknown and last one pulled from a phase IV, but most likely pulled from a ripper guitar with white bobbins. Seller let me borrow last one to test which had a faulty wire that I repaired and analyzed.
One difference I found was the baseplate. The baseplate on the early model is much different. There are more holes in it. Also the early model has and extra bar for holding the slugs. The baselplate is slightly bigger,compared to a standard paf. It has some extra metal where the 2 screws are. I also noticed the wiring of the pickup is slightly different, which I’ll show below when trying to clone.
Achieving Pickup Nirvana Test 1- Cloning Early Univox Phase 3 humbuckers- Matching DC, Inductance and Resonant Peak
First thing I tried to do was get in the ballpark by matching the DC resistance.
I used coil estimator jdguitarworks.com/coil/coil.html
First test – 7.81k 84.6pf capacitance, Still off, more similar to the later phase III resonant peak
2nd test 7.646k 62pf capacitance, Still off, resonant peak raised a little but the DC resistance lowered .17k and capacitance lowered. I wired this one like the Univox. The other start ground was wired directly to baseplate ground, instead of separately like above. This allows me more headroom to add more winds and keep the DC resistance similar to get that inductance higher and closer to 2.99 kHz target range.
Might also try some other things to get that inductance higher. Will add some sheet metal strips to baseplate to try to get that extra bar they added. Hopefully this will get me closer. Also adding the pickup cover will lower the resonant peak about .04 kHz. Obviously the headroom lets me add more winds also.
Notes below and guide for setting up presets on Mojotone Winder. Yes Baby powder is the magic to not snap coil wires. Test 5 I’ll try more winds and I ordered some 1/16″ sheet metal to add to baseplate to mimic the effect it has on eddy currents. Also order some thinner 28 gauge lead wires. Google Sheet of my results below also cut and paste some of Antigua’s data below to compare. There are some pickups that are close to the early univox one, Lollar low wind imperials and Seymour Duncan Anitiqua’s, but those are really pricey.
SAVE PRESET
Program>Down=
Press “UP” for CC
Program>Turns
Program> Preset “UP”-Yes
Press “UP” or “DOWN” to run through presets
Find preset Press “Program” to select preset
Press “Up” to save preset
LOAD PRESET
Program>Down=
Press “UP” for CC
Program button until “Use presets” shows
Press “UP” or “DOWN” to find “Restore Preset 1-10”
Find preset Press “Program” to select preset
Press “UP” to load preset
Press “RUN STOP” to start winding
Preset 1 4333
Preset 2 4888
Preset 3 4955
Preset 4 4940
Preset 5 4981 TBD
4900= 4350 = 8.7k= first pickup at venn, wind count not accurate?
4900/8.7=x/7.7 incorrect resistance calculated
X=563*7.7
X= 4336
correction
4900/4.35=x/3.865
x=4353 which doesn’t equal 3.865 from first test
1st test -destroyed for 2nd test
4333 turns 3.37k = 6.74k
4333/3.37=x/3.865
x=4969 estimate to get target 3.865k
semi accurate according to coil estimator for tight scatter recommending 4889, will try on preset 2
2nd test-destroyed for 3rd test
4888 turns 3.822k =7.64k
4888/3.822=x/3.865
x=4943 estimate to get target 3.865k
Average last 2 estimates 4955
will keep this coil and match with 4955
3rd test
4955 turns 3.9k = 7.8k
will mix with 4th test and analyze should equal 7.722k
Test 4
4940 winds
with test 3 assembled in humbucker, 3.13kHz resonant peak
Need more inductance, steel poles? add cover.
Test 5 4981 turns the dc offset is -.17k, target 7.73k to 7.9k/2= 3.95k target
Coil estimator estimates 4981 turns to get to 3.95k target. Assuming the grounding to baseplate has similar lowered DC effect.
Full results of 6 univox pickups some without covers and baseplates. Baseplates have major impact on eddy currents.