| |
G8MNY > TECH 15.02.16 12:32l 142 Lines 6689 Bytes #999 (0) @ WW
BID : 60264_GB7CIP
Read: DK3UZ GUEST OE7FMI
Subj: Marconi Counter Type 2432A
Path: DB0FHN<DB0PM<OE5XBL<F1OYP<IZ3LSV<ED1ZAC<CX2SA<GB7CIP
Sent: 160215/1014Z @:GB7CIP.#32.GBR.EURO #:60264 [Caterham Surrey GBR]
From: G8MNY@GB7CIP.#32.GBR.EURO
To : TECH@WW
By G8MNY (New Jun 13)
(8 Bit ASCII graphics use code page 437 or 850, Terminal Font)
I have repaired one of these old Marconi Instruments 560MHz Precision Frequency
Counters, Model No.52432-313Z.
Not working, was no -12V, caused by a shorted tantalum capacitor had taken out
the T05 pass transistor. The +12V +5V rails were both dangerously high, due to
no negative rail on their control 741 op-amps!
旼컴컴컴컴컴컴컴컴컴컴컴컴컴컴컴� >1 50MHz-560MHz, 50mV-5V max, 50� BNC
� _ __________________ � Burst Mode, holds reading on carrier loss
� (_) 놜8.8.8.8.8.8.8.8.:� � >2 10Hz-200MHz 25mV-5V (250V 50Hz) 1M� BNC
�� 읕컴컴컴컴컴컴컴컴켸 틥 AF filter, <200kHz
絹 _ _ 틜 LF Multiplier, Res 0.01Hz to 13kHz (1Sec)
볐 (_) � � � (_) � � � � � 낳 Gate times 10S, 1S, 10mS, 1mS
볐 1 2 낳 Resolution 0.1Hz,1Hz, 10Hz, 100Hz �1 count
� ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ � Oven Ref Output & Ext 1MHz 1V BNCs
훤袴袴袴袴袴袴袴袴袴袴袴袴袴袴袴袴� Supply 110/120/220/240V AC IEC.
A simple fault, if none of the chips were fried with the over voltage, it
should work now? But life is never that simple, as that was not the only fault!
The 1S gate range read 10x too low & other ranges occasionally read 0.6x low.
I could not find a diagram for this "A" version, the board layout & circuit is
different to the 2432 diagram I did get. The 3 power suplies are all different,
but the counter ICs are the same, so a starting point.
Sequence
Drivers Timers 40pin IC TOP PCB Up End Out Rigger
=============================================== Connection
FRONT �7seg Joining읓쩟쩟쬔 Hing out
� PCB|||||| /~~~~~~~~~~\ [ MAINS IEC
�LEDs 旼||||||커 � Mains � [ & Volts
_ _ _ _ _ _ _ XTAL� |||||| � � Toroid � 커 REAR
BNCs Screened OVEN� |||||| � \__________/ 켸Ref BNCs
IN 旼複 Pre-amps UNIT넬좋좋좋엿 metal frame oCal
& SW읕賽=============================================
- - - - - - - Pre-scalers * PSUs BOTTOM PCB
Pressing hard on the top PCB did change the faults, but a good solder up over
the 2 PCBs, did nothing! All these faults eventually proved to the large main
special counter IC, with it's 40 gold plated pins & socket on the top PCB.
Re-seating it MANY TIMES initialy cleared all the faults, so no blown chips,
but they did not stay away for long! Looking inside the IC holder there was
several broken socket fingers. So I soldered the IC directly in. (well you
would never get a replacement IC)
HOW IT WORKS
The 2 inputs both see pre-amps with a common AGC (90dB LF, 54dB UHF) & are
relay selected, to feed a special gated UHF �10 pre-scaler IC, then to an HF
�10 pre-scaler, & to a �2. The resultant �200 feeds the slower special main 5.5
digit counter IC. The pre-scalers also feed 9 BCD data lines read during gate
reset, for the additional (LSD) 2.5 digit information to make 8 digits in all.
TB Gate Gate time select
旼컴<컴컴컴컴컴컴커10m100m1S10S LF
� Resets � v v v v v
VHF-UHF � 旼컴컫컴�<커� |..|..|..|..|......DP......
1)fuse-Preamp__ 旼좔커旼좔커憫커 憫좔좔컨컴좔컨컴좆 \
혽GC \_놔10 척�10 척�2쳐>� MAIN 5.5 DIGIT 쳐>DRIVERS->8x 7seg
2)컴x컴Preamp__ 읓쩟쬔읓쩟쬔읕쬔�>� COUNTER IC � LED display
AF LF-VHF A\\\__B\\\__C\_� 읕컴컴컫컴컴컴쩡쬔
Filter BCD data � | |
� | �-->Overflow LED
Ref 10MHz컴>퀀10컫컴\__�10컴>컴�Ref |
Osc Oven � � 100kHz �-------->Gate LED
Int)<컴컴컴컴컴컴컴컴컴컴컴켸 쿐xt Ref
1MHz 쿌uto------------------------>Ext LED
Ext)>컴컴컴컴컴컴컴컴컴컴컴컴켸Sence
The main IC determines the TimeBase gate time & is feed with 100kHz reference.
This is derived from a fairly accurate & stable 10MHz Xtal in an electronic
controlled oven.
The fast reading LF multiplier range (a useful LF feature) gives resolution
down to 0.01Hz. It counts reference clock for 1 input cycle (100kHz Ref & Gate
swapped) & then the reciprical is calculated & displayed. But it only works up
to 13kHz!
TESTING & CALIBRATION
The +5V rail has a preset for setting it to +5.10V. The +12V rail follows, &
the -12V is independant.
Input 1 VHF-UHF, pre-amp tends to oscillate @ 400MHz with no input, & a lockout
preset (like a squelch) kills the erroneous reading. when adjusted well it
managed >810MHz @ 1V input (5V max=250mW).
Input 2 AF-VHF, pre-amp has a Schmitt trigger, & setting the bias preset (DC1)
well, is very important for best sensitivity, & good VHF. It managed >320MHz @
1V input.
Both pre-amps share the same AGC preset with 6V DC set testpoint (no signal).
Reference oscillator calibration is done after at least 15 min warm up with the
rear multi-turn preset. Compare the 1MHz output with an off air locked 1MHz on
dual beam scope (while counting something), or just count a reference source.
Overflow LED lights over 99999999 count, but the display is still accurate for
>8 digits.
OVEN STABILITY
Square law Time-Hz plot of a 10MHz off air locked reference.
Hz _
+20� � '� Fast Oven
+10� � ' Heat Up Heater &
5.0� � ' Overshoot Thermisitor
2.0� � ' Delay Oven Cycle
1.0� � ' .�''''�. .頰�.
0.5� � ' �' '''' '�...�''�. ..
0.2�_�_ _ _ _� _ _ _.'_ _ _ _ _ _ _ _ _ _ ''_ '�''�.�'頰�..
-0.2� � , .
-0.5� � � Longer term Oven & PCB warm up heat drift
-1.0� � '_.�'
-2.0� �
-5.0� �
-10달
-20달
-50닫
-100닫____________________________________________________________________
0 0.5 1 2 4 8 16 30 60 120Min
In use at room temperature it gives repeatable accuracy of < 2/10^7 �1, or
less than �2Hz in 10MHz after 1 min warm up! But 1 hr for 20x better stable
reading of < 1/10^8 �1, or less than �0.1Hz in 10MHz!
The internal reference is 10-100x more stable/accurate than a non-ovened Xtal
in a cheaper counter, or for that matter your rig !
See my other tech buls "Watson 'Hunter' Counter Mods", "PIC Freq Counter Mods",
"Old Venner Counter Type TSA3334", "198kHz Off Air Standard", "Off Air Lock for
Ref Osc" & "Calibrating Frequency".
Why Don't U send an interesting bul?
73 De John, G8MNY @ GB7CIP
Read previous mail | Read next mail
| |