Parallel HB Vs Series HB output

[cooltone]

Hi,

I have not been on for a while, but I am still working on a design (at a snail's pace).

I would appreciate your comments on this view of the comparative voltage output of a parallel vs series HB.
The conclusion of this piece is that the voltage output is the same whether connected in parallel or series.

Assumption 1:
The kinetic energy (power) from the strings is coupled into the coils via the magnetics.

Assumption 2:
The amount of coupled power into each individual coil is not changed whether the coils are connected in series or parallel

Assumption 3:
All power coupled into the coils must be dissipated into the resistive elements of the circuit

The total coupled power in both coils = P (whether series or coupled)

Since P=V^2/R, then V=SQRT(P/R)

Then Vs/Vp = SQRT(P/Rs)/SQRT(P/Rp) = SQRT(Rp/Rs)

Where
Rs = Resistance of coils in series + Total resistance of VOL//Tone pots
Rp = Resistance of coils in parallel + Total resistance of VOL//Tone pots

Since Total resistance of VOL//Tone pots >> Resistance of coils ( in series or parallel) Rp/Rs is very close to 1
So Vs/Vp is very close to 1 - in other words the voltage output is the same

What do you think? Have I got something wrong here?

Many thanks for your comments,

Cooltone

[col]

Hi Cooltone,

I'm weak on theory, but it is generally accepted that you will have a higher output when pickups/coils are in series. If you think about it, output is relevant to the reference point (ground). So, if you have a second pickup/coil, in series, its reference point is already above ground because of the signal generated by the first coil. If we are talking about two pickups (rather than two coils), then the signal strength will not be doubled, because but there will be some harmonic cancellations between the two pickups (but still a very strong peak signal). But, with two coils within a single pickup: I would expect that most of the output/voltage (across the frequency range) will be nearly doubled. Again, I am weak on theory, so someone else will explain it better than me (or even correct everything I have written).


Edit: typo.

[col]

Oh, I should have added, there will be some filtering effects too, but they will mostly affect higher frequencies.

[ashcatlt]

It's both theoretically and empirically false. There's no question about it. I do believe that if you consider the current grnrrated by the movement of the magnetic field as constant, then the voltage across the coils is proportional to the resistance through them, which is 4 times bigger when series connected than in parallel. This is knocked back down a little by the voltage divider created by the resistance of the coils over that of the load, but since the load is usually much larger than the coils, the proportional difference there is smallish.

Look at it another way: two 9V batteries in series will give you 18V, but in parallel they will only give you essentially 4.5 (9V divided down by a 1:1 divider) times 2, which is only 9V.

To put it another way - in series, the bottom coil pushes the bottom of the top coil, which then pushes on top of that for itself, whereas the parallel coils are both kind of pushing up from the same ground, but they're also kind of dragging each other down at the same time.

You might try to argue against any of that, but it doesn't change the fact that if I flip my S/P switch, the meters show real differences.

[cooltone]

Hi Ashcalt,

That is what I thought, but now I am not so sure.

If a constant current is generated in the coil and the coil voltage is generated across the coil resistance, the same current must complete the circuit in order to flow and this means it must go through the resistance external to the coil. Typically the external resistance is much bigger than coil resistance, I cannot see how the coil voltage is defined by coil current x coil resistance.

This chap did some measurements:
www.moore.org.au/pick/06/06_gobd.htm

He shows some measurements (graph) comparing parallel/series connected HBs and writes:

"The second pair of curves is particularly interesting as these are the hum bucker connected as a series (HMB01SS) and as a parallel (HMB01PP). Both have exactly the same sensitivity as each other, and also about 8 dB less than the single arrangement."

All this suggests the voltage is developed across the coil inductance and not the coil resistance, which means the battery analogy may not be valid.

I am only asking this because it means the choice of tone and volume resistance has quite an impact on the circuit I am looking at.

[JohnH]

That is an interestinf article. But there is something amis where it seems to suggest that series and parallel wiring have equal sensitivity. Its not true, we can hear, measure and explain how series is louder!

[ashcatlt]

If I plug my hybrid into the instrument input on my interface, turn off all but the mid pickup, and put it in parallel mode, I can crank on the strings as hard as I want without quite hitting 0dbfs. Switch it to series and it clips like crazy. That is all the proof I need. Perhaps my theoretical explanations are not quite correct, but the proof is in the jello.

[JohnH]

Also, if you are scheming designs and wish to explore this analytically, the GuitarFreak spreadsheet now has most of this covered:

GuitarFreak

The latest release has nearly all the electrical, string vibration and pickup response physics built it, and will deal with pickup position, series/parallel, phase reverse, humbucker/single, picking position etc, and show you the results visually. (There are some magnetic subtleties not captured, but the width of string pickup sensing is allowed for)

As an aside, I just ran some inductance measurements on a bridge humbucker on my Strat. I got readings as follows:
full series 4.7H,
split single 2.3H,
parallel 1.2H.

All fairly close to as expected from simple tbeory, given my meter is not high-end, and the volume pot was still in circuit.

[cooltone]

You both pushed me to re-examine the situation.

You are right and my assumptions are incorrect.

I realise that assumption 2 is false. If the coil is open circuit (infinite impedance), no power dissipation. If coil is short circuit, all power is dissipated in the coil resistance. So power transfer from the magnetic circuit to the electrical circuit must also be variable.

I agree something must be amiss with the 'moore' article.

JohnH, many thanks for letting me know about GuitarFreak. For the moment, I will stick with LTSpice, I have put a lot of effort into learning it and developing a multi-pole, multi-throw switch model, and I am close to finalising the circuit.

FYI: I will be using Artec GVH 59 HBs. I measured the parallel self-inductance over three resonance points for two sets of pickups using an oscilloscope and a tone generator and got:

Bridge
L: 0.96H
C: 220pF
R: 2.07k

Neck
L: 0.808H
C: 220pF
R: 1.91k

The L looks low compared to typical HBs, but it is correct.With a 400pF amplifier/cable load, I am expecting a very 'acoustic' sound. For other tones, I will switch-in capacitors to set different resonances.

(NB: The tone generator was rubbish, but enough to do the job. it was a mobile app and generated to much HF noise from the phone).

Cheers,

Cooltone.