John Mihevic On The Goff Capacitor Kit

This is a review of the Goff _tone generator capacitor kit_ which was contributed by JohnMihevic. (See ToneGeneratorCapacitorReplacement for more on the subject.)

My 1956 B3 (Ser # 59520) did not have much "brightness" in the higher frequencies. I was always pulling out all the upper drawbars to get more highs. I decided to replace the wax capacitors that are part of an LC resonant filter for frequencies 49-91. I purchased the Goff capacitor kit and am very pleased with the results. There is much more clarity and brilliance in the mid to upper frequencies with increased drawbar control of tone. The volume output is increased in that range and the percussion has more "bite". Now each drawbar has more definition and each step change on the upper drawbars is more noticeable than before. I have seen some complaints that the new Goff recaps can sound too sharp and that some prefer the warm sound of the older organs with the waxed caps. I don't agree. If I want, now my B3 will really scream now or I can cut back on the drawbars and get the warm sound that was always there, it is like having the best of both worlds.

The recap job is straightforward, just time consuming. Replace one component at a time. Easiest access to the Tone Generator (TG) is from the front of the organ by lifting both manuals and supporting the front corners of the lower manual with wood blocks. I would also recommend sliding a piece of thin cardboard on either side of the manuals to prevent scratching the organ case should the manuals slide sideways. Trek II makes a pair of case scratch protectors specifically for this purpose.

The factory capacitors for frequencies 49-54 was 0.255uF and 0.105uF for frequencies 55-91. The waxed capacitors had increased in value through the years. I measured the existing capacitors and found one 0.255uF had increased to 0.590uF. A 0.105uF had increased to 0.540uF. I also measured mv RMS values before and after the recap. All mv outputs increased. Some outputs more than doubled. An example was TG freq #78 which increased from 2.96mv to 6.45mv.

The Goff capacitor kit replaces the 0.255uF capacitors with 0.224uF and the 0.105uF with 0.104uF. I checked the "tuned" value of the new capacitors by using Sprague "orange drop" capacitors to increase/decrease values from the supplied capacitors. The 0.104uF were matched very well to the reactors with peak output for the frequencies 55-91. However, I added a 0.015uF capacitor in parallel with the 0.224uF capacitors. The 0.015uF capacitor produced the best peak output for my TG with an increased output of 0.5mv to 1.0mv for frequencies 49-54. The typical output for these frequencies was 4.3mv before the parallel capacitor addition.

The Goff recap kit also comes with R/C filter networks for frequencies 37 to 48. They were factory installed on B3's and C3's after around ser # 93,000. Goff says these were to reduce crosstalk and 60Hz hum for these frequencies. Supposedly this occurred when the first brown drawbar was pulled out with other drawbars. Since this is an R/C passive filter, adding the filter will reduce the TG output. I don't have hum or crosstalk that I can hear and I didn't want any reduction in TG output, so I decided not to install these R/C filters.

I would highly recommend a TG recap for anyone with the older wax capacitors. It certainly puts new life into the old Hammond.

I don't think the Goff capacitors are .224uF and .104uF respectively. "224" and "104" are the markings on the capacitors, but these are interpreted as 22x10⁴pF and 10x10⁴pF respectively. In other words, 0.22uF and 0.1uF. There are no such things as 0.224uF and 0.104uF capacitors. -- StefanVorkoetter

Capacitor Matching & R/C Filter Networks--Update on Tone Generator Capacitor Replacement--John Mihevic 01/23/2011

First, thanks to Stephan Vorkoetter for noting that the correct values for the Goff capacitors is 0.22uF and 0.1uF

The tone generator capacitor replacement I did in 2005 (described above) restores the tone generator performance to 90% of a new factory generator manufactured by Hammond in the '60s. To get the tone generator equivalent to Hammond's performance when new, I installed the R/C filter networks for frequencies 37 to 48 and individually matched capacitors to each filter coil for frequencies 49 to 91. Both of these improvements are described below.

Addition of R/C Networks

I was not satisfied installing the Goff R/C filter networks for frequencies 37 to 48 without adjusting the coil magnet assembly for each frequency. The R/C networks load these frequencies resulting in a reduction in volume. The output was reduced by 1.5 mV - 2.0 mV RMS out of a nominal 4.5 mV, My solution was to measure the output of these frequencies before R/C installation, install the R/C networks, and adjust the coil magnet assembly closer to the tonewheel to increase the mV output of each coil to the values recorded prior to R/C installation. I am confident Hammond calibrated the tone generator with the R/C networks installed so the filter loading was part of the calibration.

Capacitor to Filter Coil Matching

The capacitors in the Goff cap kit are matched in value to each other (0.22uF and 0.1uF). However, they are not matched to the filter coils as was done at the Hammond factory. The following paragraph is taken from a Hammond tech manual I have from the late 1960's with the requirement to match each coil/capacitor for frequencies 49 t0 51. This is in Section 5 (Main Generator), page 3 of the manual.

      Two condenser values are used - 0.255 mfd for frequencies 49 to 54, and 0.105 mfd for frequencies 55 to 91.  The transformers are all different.  Each transformer is matched to its condenser and any replacements are supplied as matched pairs by the factory. 

For me to accomplish this resonant frequency matching, I purchased +/-10% Sprague Orange drop series 225P capacitors from Newark Electronics. It was the +/-10% tolerance that allowed me to match each capacitor to each filter coil. The values I used were .22uF(220nF) .082uF(82nF) .015uF(15nF) .01uF(10nF) .0047uF(4.7nF) .0033mF(3.3nF). Note that 1000nF = 1uF The actual range of measured values were:

<pre>
 Marked Value       Actual Range In Value

   220.0nF           230.0nF – 241.0nF
    82.0nF            83.2nF -  91.4nF
    15.0nF            15.40nF
    10.0nF             9.50nF – 10.18nF
     4.7nF             4.86nF – 05.21nF
     3.3nF             3.60nF
</pre>

In many cases I used two capacitors in parallel to get the correct value. The final value of capacitance was obtained by increasing the value of capacitance until the maximum mV output was obtained. More or less capacitance reduced the mV output. The model of a real capacitor includes a resistive and an inductive component. The magnetic pickup has resistance and inductance. All of these, and some components that I don’t even know about, affect the resonant frequency which is the desired output of a given tonewheel. The equivalent circuit/math model is complex. With my method, tuning is done with the actual capacitor(s) that will be soldered to the filter coil. Every circuit component is included in the capacitor/coil matching. In this way a tuned circuit resonant peak is obtained at each tonewheel frequency with optimum filtering of unwanted frequencies and the signal to noise ratio is a maximum for that frequency. I should mention that I still have enough “crosstalk” that is part of the classic Hammond sound. All work was done with both manuals raised and the tone generator in full operation. Access to the tone generator was from the front of the organ. The table below has the final value of capacitance that resulted in the maximum mV output for each frequency from 49-91. Column 1 is the “Frequency No.” Column 2 is “Capacitor(s) Used”. Column 3 is the “Total Capacitance” value. Column 4 is the value of the “Goff Cap Removed”. Column 5 is the value of “Wax Cap” that was originally installed by Hammond.

Freq   Capacitor(s)    Total    Goff Cap  Original
No.    Used            Cap      Removed   Wax Cap
<pre>
       nF              nF       nF        nF
</pre>

49     241.0   -       241.0    225.0     580
50     233.0   -       233.0    225.0     412
51     230.0   -       230.0    226.0     400
52     230.0  9.90     239.9    226.0     485
53     237.0  3.60     240.6    227.0     590
54     236.0  4.86     240.9    228.0     360
55      83.6  4.99      88.6     99.6     203
56      86.2  4.93      91.1     99.9     330
57      86.4  4.95      91.4     99.3     183
58      88.2  5.21      93.4     98.6     340
59      87.1  4.88      92.0    100.0     320
60      91.4   -        91.4    100.0     350
61      88.3   -        88.3     99.9     280
62      89.3   -        89.3    100.0     330
63      86.9  4.88      91.8     99.9     188
64      85.6  4.95      90.6    100.3     410
65      88.0  4.86      92.9    100.9     365
66      88.0  3.60      91.6    102.0     260
67      83.2  5.13      88.3    100.5     350
68      87.6  4.94      92.5    101.2     290
69      85.6  9.83      95.4    100.8     238
70      88.6  9.68      98.3    100.5     320
71      87.7 10.03      97.7    101.5     400
72      85.9  5.00      90.9    102.0     320
73      88.6   -        88.6    102.2     340
74      85.2  4.98      90.2    101.8     215
75      90.2   -        90.2    101.9     360
76      87.3  5.00      92.3    102.1     270
77      90.2   -        90.2    101.8     223
78      85.6  9.85      95.5    101.8     330
79      87.5  9.50      97.0    102.3     215
80      84.8  9.57      94.4    101.6     315
81      88.1  9.78      97.9    103.5     300
82      84.8 15.40     100.2    103.5     211
83      86.7  9.83      96.5    102.5     410
84      87.8  9.69      97.5    103.5     330
85      86.4  9.76      96.2    102.5     165
86      87.9  9.50      97.4    102.5     322
87      88.9  9.76      98.7    103.6     410
88      85.9 10.18      96.1    103.0     184
89      85.2  9.56      94.8    103.4     540
90      87.9  4.96      92.9    104.2     380
91      88.0  4.98      93.0    105.6     250

At every frequency, the capacitor to coil matching produced a higher millivolt output than with the Goff capacitors that I replaced. The highest change was +1.72 mV. The drawbar definition is increased with more crispness, clarity and presence of the organ tone/timbre. Capacitor/coil matching is not for everyone (it is time consuming), but if you are so inclined, the reward of restoration, identical to what came out of the Hammond factory, is well worth it.

—John Mihevic