Вставка
- Опубліковано 11 лип 2024
- This video shows the teardown, the reverse engineering, the repair and the test of a vintage Cessna dual turbine RPM indicator.
00:00 - Intro
00:37 - Teardown
03:04 - Preparation for reverse engineering
04:17 - Reverse engineering part 1: power supply
05:33 - Reverse engineering part 2: main circuits
07:20 - Issue #1: wrong transistor orientation
09:15 - Issue #2: cracked solder joint
10:11 - Test
Note about the construction: the electronic side of this indicator was poorly constructed. The dual layer non-plated through hole PCB without solder mask nor conformal coating has resulted in a solder joint crack. Someone attempted to repair this instrument (maybe because of this issue) and one 4-pins Darlington transistor was not fitted correctly during this repair.
Internal circuits: the input signal is fed to a comparator which delivers a square wave signal. This signal is fed to a discrete monostable which delivers a rectangular signal. The duty cycle and therefore the mean value is proportional to the input frequency. After filtering the DC voltage created is used as the input of a servo control loop. The feedback device is a potentiometer. A chopper using two dual emitter pnp chopper transistors converts the DC error voltage into an AC error voltage which is fed to the servo amplifier which drives the AC motor.
This instrument incorporates an inverter for each channel for the generation of the AC voltage required by the motor.
The use of an AC motor and the choice of a DC power supply voltage complicate dramatically the circuits of this instrument.
Particular parts used:
Dual emitter PNP chopper transistor: Sprague 460070-2
Servo amplifier power stage and inverter: Darlington U2T101
Pinout from reverse engineering:
(x): left engine
A (K): input signal (5Vpp min, positive signal), 70Hz for 100%
B (F): common line for power supply and input
C (E): positive power supply (14 or 28V, automatic selection by the unit)
About bipolar chopper transistors and circuits:
www.bitsavers.org/components/m...
Recorded 2023/07/23 - Наука та технологія
I agree with you on the quality. I've never seen a PC board in that class of avionics so terrible. I've seen poor quality in general aviation class avionics because they have to be affordable. But this was a business jet class instrument. The lack of conformal coating on the boards is inexcusable as is the method of wiring parts to the board. I can't see where they saved money since the assemblers had to manually route component wiring to the traces. As always, a fantastic reverse engineering job! Best on UA-cam!
The revers engineering is an immense work..
As always awesome work and explanation..
Thanks Michel.
Thank you for showing us the process and some of your intermediate steps in reverse engineering this.
really enjoy your reverse engineering work 👍 great skill
Hello, oui la qualité semble vraiment pourrave... bizzare pour de l'avionique, les standards sont pourtant assez hauts.
That was a lot of work. Well done.
This design however is terrible. Not just all the mechanical problems you pointed out but also the design:
1) First I did not understand how the power supply could even start oscillating. Then I concluded that the rise time of the power supply input voltage would be enough to make the transformer generate it first output impulse. So when then input voltage rises slowly there is a chance that the power supply does not even start oscillating. Bad design.
2) The monostable multivibrator: Once Q4 conducts, the fully charged capacitor would pull the base of Q3 down to about 1.4V-VA. If VA, i.e. that Zener Diode is more than 7.5 Volts or so I would be worried that the maximum allowed negative base voltage of Q3 is exceeded. Also when this Zener diode breaks open then transistor could be damaged.
I don't like the design. But your reverse engineering was top notch!
Isso é que eram máquinas
Whilst I like the design, I'm astonished that the low quality of this workmanship would pass for an avionic instrument! It's only fit for a child's toy, at best!