X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 1
Hammond
X-66
Renovating the
Vibrato Driver Preamp
by Dan.Vigin
Hammond® and Leslie® are registered trademarks of Suzuki Musical Instrument Manufacturing Co. Ltd.
Binche / Belgium Dec.2010
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 2
TABLE OF CONTENTS
- Forewords & Purpose of this project P. 3
- General Description P. 4
Original Vibrato Driver Preamp circuit
- X-66 Headroom concept P. 6
- New VDP circuit P. 8
- Transistor MJE340 P. 9
- Emitter resistor P. 10
- Schematic Diagram of new VDP P. 11
- Output levels P. 12
- Realization P. 12
- Conclusions P. 12
- PCB Layouts P. 13
- Datasheet MJE340 P. 14
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 3
Renovating the Vibrato Driver Preamp on Hammond X-66.
Forewords – Purpose of this Project
Several weeks ago, when playing an abnormal level of background noise and hiss was
detected and this excessive noise made me nervous enough to start investigations in
order to reduce this annoying noise.
This background noise was mostly audible in Drawbars use while being less significant
with either Tabs or Percussions.
Even with no key depressed, by checking the outputs of the Vibrato Driver Preamp
board, with the help of an oscilloscope, I found at once that one channel exhibited
much more noise than the two others. This faulty channel was the Drawbar channel.
So, I decided to pursue my investigations. Measured voltages were normal, capacitors
were previously replaced by new ones.
The conclusion is rather simple, the driver transistor creates this inherent noise in the
Drawbar channel.
The first reaction is to find and replace this faulty component by a new one.
Unfortunately, this driver transistor labelled 21221 in the Hammond parts list is the
equivalent of RCA 39699. This transistor was produced in the early 60's and is in the
line with the first generation of Si NPN transistors launched at that time by RCA, about
50 years ago !.
Needless to say that this component is now unobtainable, at least in Europe.
In view of this situation, I decided to redesign this board and re-install much updated
transistors providing better performances than the former RCA 396999.
This is exactly the purpose of this chapter.
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 4
General Description.
Original Vibrato Driver Preamp circuit.
The board AO-32297-1 is located at the back side of the scanner block, under the
Upper keyboard.
Personnaly, I don't like at all the way this board has been engineered. This is the kind
of board commonly used to create prototypes in lab's but not for production units.
This is a personal opinion.
Top view
Bottom view
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 5
The three Vibrato Driver Preamp's ( VDP's for short ) are located on the same board.
The concept of each VDP is the same however they all have some slight differences
that have to be taken into consideration.
In fact, the role of each VDP is to procure enought amplitude to 'drive' its
corresponding vibrato phase shift line that is linked directly to the scanner.
The VDP board is then composed of three sections:
- on the left, the TREBLE VIBRATO DRIVE PPREAMP
- in the center, the TAB VOICE VIBRATO DRIVE PREAMP
- on the right, the CELESTE VIBRATO DRIVE PREAMP
There is no so much to tell about the circuitry. Basic 'Common Emitter' configuration is
used with a partial decoupling in the emitter resistors.
Refer to original schematic diagram hereunder.
Fig 5-7 (partial) of X-66 Service Manual.
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 6
X-66 Headroom concept
We can notice on the diagram that those preamp's are fed by a voltage of +250 Vdc
which is totally unusual for a conventional audio preamp. In the X-66 console, this
voltage is even more, +265 Vdc were measured !
When looking on the original schematic diagram Fig 5-7, we can see that the incoming
voltage at the Treble Vibrato Drive Preamp (Drawbars) is only 0.05 Vpp (Pt. E from
Flute Filter) and the output voltage is 3.5 Vpp. So the amplification factor is 70.
Then the question is: " Why to feed this circuit with such a high voltage of +250Vdc ? "
After invesitigations, I came to the conclusion that one reason of feeding this circuit
with +250 Vdc is to procure very high "headroom" to the outgoing signal.
Headroom can be defined as ' the difference between the normal operating level and
the clipping level (or undistorted value) of an audio device'.
During tests, I found that the announced 3.5 Vpp level as indicated on Fig 5-7 can be
easily obtained with a supply voltage of +25 Vdc. However, if we increase the level of
the incoming signal, distortion appears rapidly (mostly clipping).
If the same circuit is powered with +250 Vdc (i.e. 10 times more), then clipping
(distortion) only appears when the outgoing signal reaches 115 Vpp i.s.o. 3.5 Vpp !
This way of doing provides an 'headroom' of 115 Vpp – 3.5 Vpp = 111.5 Vpp.
Generally, headroom value is expressed in dB.
In this case, the headroom is + 29,8 dB, unusually high.
Extensive measurements were conducted on workbench and as an issue the next
graphic was outlined.
X66 Vibrato Driver Preamp
0,00
20,00
40,00
60,00
80,00
100,00
120,00
140,00
24 48 100 150 200 250
Vcc (+Vdc)
Vout (Vpp)
Vout with constant input
Voltage of 50 mVpp
Vout with Vin adjusted
before clipping (Max Vout)
Headroom
(+ 29.8 dB)
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 7
The purpose of having sufficient headroom above normal operating signal is to avoid
clipping and hence audible distortion.
It is evident that when Percussion signals carrying very sharp transients are in use,
sufficient headroom is needed in order to avoid clipping.
For Tabs Voice signals, it is less needed since levels are quite constant while 'spikes'
with Kinura tab depressed f.i. may justify some adequate headroom.
For Drawbars signals issued from the flute filters, then such amount of headroom is not
at all required because signals driving this VDP are almost pure sine waves or
harmonics components of them.
The X-66 designers of this VDP do have certainly other good technical reasons which
remain unclear to me up to now.
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 8
New VDP circuit.
Since very few information as far as the specifications of this VDP board are available
in the X-66 Service Manual, the original board AO-32297-1 board was dismounted and
put on the workbench for test.
As stated earlier, only one channel was defective (noisy), so the two others were
workable for measurements.
The original board AO-32297-1 was powered by +250 Vdc with an incoming signal of
50 mVpp @ 1Khz . The collector of transistor 21221 (RCA 39699) was loaded by a
12K/1W resistor to ground via one 1µF/400Vdc capacitor. See simplified schematic
herafter.
The three sections of the AO-32297-1 board were configurated so and results of
measurements were recorded in order to redesign a new VDP board.
With the help of a workbench simulator software, a new VDP circuit was redesigned
based on MJE340 transistor having much superior performances than RCA 36699.
VIBRATO
DRIVE PREAMP
12 K /5 W
82 O33 O
22 K
E
C
B
+
-
-
+
MJE340
100 µF
63 V
1 µF
63 V
1 nF
220 K
T1
E
B
C
2
1
3
TO-126
MJE340
Test-Point
1 µF
400 V
GND
50 mVpp
@ 1KHz
12K/1W
+ 250 Vdc
Load resistor
applied during
measurements
IN
OUT
To Vibrato Phase
Shift Line
Foot emitter
resistor
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 9
The overall new schematic diagram can be found on the next page.
This new schematic diagram is almost identical to the original one. Beside the
replacement of the old RCA transistor by MJE340, only the collector resistor has been
changed from 10K/5W into 12K/5W to slightly decrease the collector current and
reduce to a certain extend the power dissipation that is not negligeable due to high B+
voltage involved. Dale 12K / 5W – 1% resistors were installed.
Here is the new VDP board.
Transistors MJE340.
While not necessary by calculation, for reliability purpose, the transistors MJE340 were
mounted on heatsink. Silicon compound was used for better heat transfert.
It is strongly recommended to select hFE values (DC current gain) of those transistors.
hFE may vary from 30 up to 240 according to datasheet of the transistor supplier which
is indeed a very wide range.
The correct hFE value for each MJE340 transistor should remain in the range of min. 70
to max. 85. In my case, the three transistors had an hFE value of 83.
Datasheet of MJE340 are also provided in the next pages.
Simplified characteristics
RCA 39699 MJE340
NPN Technology Early generation Si Transistor Planar Epitaxial Silicon transistor
VCB0 400 V 300 V
VCE0 300 V 300 V
VEB0 6 V 5 V
Collector Dissip. 8 W 20 W
hFE 40 ~ 80 30 ~ 240
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 10
33 ΩΩΩΩ Emitter resistors.
In addition to that, if one wants to keep an equivalent output level at each section of
the VDP board, it is also important to carefully select the foot emitter resistor of 33 Ω.
Higher value of this 33Ω resistor will decrease the overall gain while lower value will
increase the overall gain.
Note: during life-tests, for reliability purpose, the voltage supply was increased from
+250 Vdc up to +270 Vdc. No breakdown was recorded.
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 11
TREBLE VIBRATO
DRIVE PREAMP
12 K /5 W
82 O33 O
22 K
E
C
B
+
-
-
+
MJE340
20 µF
63 V
1 µF
63 V
12 K /5 W
82 O33 O
22 K
E
C
B
+
-
-
+
MJE340
100 µF
63 V
10 µF
63 V
12 K /5 W
82 O33 O
22 K
E
C
B
+
-
-
+
MJE340
100 µF
63 V
10 µF
63 V
1 nF
1 nF 1 nF
+ 250 Vdc
TAB VOICE VIBRATO
DRIVE PREAMP
To Expression
Preamp Board
CELESTE DRIVE
PREAMP
GND
220 K 220 K 220 K
20 nF
400V
T1
W46
T2
T3
0.1 µF
400V
1 µF
400 V
1 µF
400 V
+ 250 Vdc
33 K
47 K
W55
W65
W64
W57
To Drawbar
Vibrato Phase
Shift Line
To Tab Voice
Vibrato Phase
Shift Line
To Celeste
Vibrato Phase
Shift Line
E
From Pedal
Preamp
Fig 5-7, X-182
G
From Percussion Section
Fig 5-4, X-76
From Drawbar
Section
Fig 5-1, T-64
Circuit Diagram – Hammond X-66
Re-designed Vibrato Driver Preamp
File: Vibrato_Driver_V1.vsd
Date : 30/11/2010
Issued by : Dan.Vigin
Hammond® and Leslie® are registered trademarks of Suzuki Musical
Instrument Manufacturing Co. Ltd.
E
B
C
2
1
3
TO-126
MJE340
+ 60.5
Vdc
+ 2.4
Vdc
+ 1.8 Vdc
+ 0.5 Vdc
8.1 Vpp
(1)
(@ 1KHz)
DALE
50 mVpp
(@ 1KHz)
16 mAdc
48 mAdc
WHT/BLU
YEL
BLK
Shielded
cable
GND
BRN
2x BLK
RED
RED ORG
YEL
47 K
Notes.
- (1) Resistor of 12 K/ 1W is used between
Test-points A, B, C and GND during
measurements on the workbench.
- (2) Carefully selected to adjust output level near 8.0 Vpp.
-
h
FE
of T1, T2, T3 to be selected between >70 and <85.
- T1,T2 & T3 mounted each on heatsink.
12 K/1W
Load resistor
during tests
AO-32297-1
Test-Point A Test-Point B
Test-Point C
(2) (2)
(2)
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 12
Output levels.
During measurements of the original board AO-32297-1, output levels of 7.2 Vpp were
recorded with an input sine wave signal of 50 mVpp @ 1KHz.
As already stated, those values are measured with a load resistor of 12K / 1W on each
section.
With the new VDP board, I have intentionally selected the transistors MJE340 and the
foot resistors of 33 Ω to obtain a value of 8.0 Vpp (i.s.o. 7.2 Vpp as on original one).
Limits for new VDP should remain around +/- 0.2 Vpp i.e. 7.8 Vpp ~ 8.2 Vpp.
This difference gives some 'voltage boost' that was positively evaluated during final
listening tests. By altering the value of this emitter resistor of 33 Ω, it is possible to
refind exactly the value of 7.2 Vpp as on the original board. It's a matter of choice.
There is no indication whatsoever in the X-66 Service Manual as far as the effective
load composed by the vibrato phase line circuits. When analysing the structure of
those circuits, it seemed to me that this load should fall in the range of 12 K.
It's a my pure guess. Measurements at no-load condition makes no sense and since
the same load value of 12 K / 1W was used for both original board and new one, the
conditions of measurements remain identical and trustworthy.
Realization.
A shown on the picture, a new board was redesigned keeping the same external
dimension of the original one. The position of the terminals was also kept for easiness
of rewiring.
The three output capacitors ( 0.1 µF/400 Vdc and 2x 1 µF/400 Vdc ) were found
perfect and hence were left in their initial position on the scanner chassis.
Negative and components sides of this PCB are provided in the next pages for those
who envisage to realize this VDP board.
There is no so much to tell about the board itself, quite basic for any qualified
technician.
Conclusion.
Initially, I had no intention to redesign this board AO-32297-1. This was caused by the
excessive 'inherent noise and hiss' that suddenly appeared in the Drawbar channel.
In fact, since I was forced to resolve that failure, from there came the idea of
redesigning this board. At the end, it was a good decision because the replacement of
old RCA transistors by MJE340's provides to the X-66 console a real 'boost' while
reducing to some extend the overall inherent background noise (rather subjective).
This is only my personal feeling.
Dan. Vigin
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 13
PCB Layouts
Components side Copper side
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 14
DATASHEET MJE340.
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 15
X66 Vibrato Driver Preamp Dec.2010
dan.vigin
Page 16
Loading...