I saw a page authour using single LM1881 above circuit to convert regular analog oscilloscope into TV monitor that inspire me to try LM1881 chip with to design the simplest TV. For more about above circuit analog oscilloscope TV monitor click here.
My friend show me from old magazine they turn the 8SJ31J Electrostatic deflector oscilloscope CRT into magnetic deflector by adding the scan coil(yoke) . I remember i have one faulty 2AP1 with the DJ1 open internally. The 2PA1 look like made around 1940s with metal base, The screen look new, is wasted to scarp, so i is good for magnetic deflection experiment.
If you also interest about DIY scope TV you can check my other page about Scope TV click here.
or Radar round magnetic deflector CRT as TV click here.
From above picture can see the internal DJ1 suppose to spot weld to the base it have open ..
DJ1 and DJ2 are a pair of deflector , I tried to leave DJ2 floating , it can pick up the hum by surrounding so i have to tight up with Anode ..
Above is the block diagram i come out in my mine, since LM1881 provide Horizontal sync and Vertical sync , i can use the horizontal Sync to drive 555 Reset pin. After all testing i found that LM1881 when there is no video signal the Composite Sync or H-sync output stay at low instead of high , so it will make the 555 shunt down when no video signal. The synchronous pulse is active low. as below
Above is initial schematic concept. I was design in direct coupling to CRT cathode the reason is i was think when no video signal the LM1881 H-sync is stay high rather then low, in order to turn off the beam for long period of time to prevent burn scree since there is no vertical sweep. but after i build i realize when no video signal it will shunt down flyback section completely. So this TV when turn on the video signal have to wait for 1minutes before the screen appear.
From datasheets there is 2 setting one for anode voltage 1000V for more brightness at same time the deflection sensitivity is decrease. another is 500V anode.. I’m using 500V anode to prolong the tube life span , the down side is the screen is more darker the dynamics of the picture is narrower.
This time i choice 13001/13003 they are cheaper and easy to find now a day and VCE(o) as high as 400V which will be good to work at 300V. 2SC1573 is old video amplifier i believe there is no manufacturing today, it can be replace with modern transistor. 13001/13003 cant replace 2SC1573 is video amplifier require to amplifier wider bandwidth with high gain, The bandwidth at least up few Mhz.
Flyback section design concept.
PCB using ExpressPCB to draw and toner transfer chemical etch.. Above is new EI28 ferrite transformer. I want to try a normal ferrite EI core see how it work in TV flyback..
Before i make my flyback transformer , i want to make sure 555 output able to synchronous with LM1881 H-sync output.
I also measure the vertical sweep output. the Vp is 5.25V and Vmin 1.04V so with this information i can design the vertical amplifier.
With the above information i can design the Vertical amplifier . DJ3 and DJ4 require 98V to deflect 1inch the screen is about 2inch so require 196V to scan full screen. Above circuit can deliver 179Vpp for input 5.2V-1V..
Primary will be 24turn. from my experiment, the peak voltage is depand the transformer self resonant frequency the higher the self resonant frequency the higher the peak voltage, so the safety capacitor bring down the resonant frequency and limit the Vpeak.. The flyback time is 4.7uS total cycle is 64uS . Due to during MOSFET turn off after flyback the another half cycle is conduct by damper diode (which is internal MOSFET diode) so i assume the total on duty cycle is 64uS minus flyback 4.7uS = 59.3uS
Boost converter Vout= (1 / 1-D) . Vin ,,,, where D = Ton/Ttotal .
D = ( 59.3uS / 64uS) =0.926
Vout = (1 / (1-0.926) ) x 12V = 163V at primary during flyback .. since primary is 24turn so 6.8V per turn.
For 600V , simply 600V /6.8V =88turn , and for 300V tapping will be 44turn.
For heater 6.3V rms = sqroot ( (sq forward voltage x forward duty cycle) +(sq flyback voltage x flyback duty))
Forward voltage per turn , 12V x 1/24turn =0.5V , duty cycle of forward 59.3uS /64uS =0.926
flyback voltage per turn 6.8V , duty cycle 4.7uS /64uS = 0.073
I remember last time i using 4 turn will try use 4 turn and calculate see i can get 6.3Vrms
sqroot [ sq(4×0.5V) x 0.926 + sq(4 x6.8)x0.073] = 7.59Vrms
try 3turn
sqroot [ sq(3×0.5V) x 0.926 + sq(3 x6.8)x0.073] = 5.7Vrms
Summary , primary 24turn , secondary 88turn with 44center tap, 4turn for heater (maybe require resistor to drop the voltage if too high)
Summary of transformer winding.
I try to power up without safety capacitor on flyback , when i slowly increase the input voltage to 5-6V , The voltage go over 200V on MOSFET Vds. So i try to add safety capacitor .
From try several value , the 10nF look good , so i able to continue increase until 12V the Vds is around 200V.
I using motorbike 12V lamp to test the heater it look good not full bright like 12V . I try to use another scope measure the Vrms it show 4.7Vrms . So i decide to incrase the 555 on duty cycle from 40uS to 50uS. (value above on PCB show is 50uS which is final value). then i able to get around 6Vrms.
The video composite signal Vpp is around 0.7V .
Due to i want to make it simple my video amplifier gain is fix, the above design show the gain is 100. later i test it the contrast is too deep to off the screen become completely dark. so i have to reduce the gain by change 1K to 3K and capacitor is 1nF for high frequency compensation. 1.5K to 4.7K to reduce the gain to 33.
I suspect the internal anode is open didnt contact well to Aquadag (carbon layer on glass) , when my deflector coil add to it it deflect the beam out of screen. so i add aluminum foil it work. i will connect to Final anode 600V to the foil.
The first tried i simply wind 20turn +20turn deflector scan coil using 0.5mm magnetic wire. it over load to my flyback . so i have to add some inductor but the current still too high, So i series the 12V motorbike lamp for current limiting.
The scan coil i series with 0.22 resistor as shunt to measure the current waveform, as expected the waveform is not linear due to the lamp act as big resistor to current limit so it make the sawtooth current compress. but from this waveform i can get the ideal how much current require to scan whole screen. The current waveform is 0.12Vpp shunt is 0.22 ohm so the current will be 0.545mApp for getting 13mm wide.
since 550mAp-p with 40turn give 13mm wide , or 24A per turn give 13mm wide.. for 50mm wide it require ( 50mm/13mm) x 24A = 92A/turn .. if i wind 300turn , the current mApp require 92A/300t =306mAp-p . So i have to try wind 300turn. which is 150t + 150t , because i not sure how many voltage require to obtain 306mApp if the result too narrow i can parallel 2 coil.
I using spray can about 6cm diameter to wind the deflector coil. I have many CCA wire , copper clan Aluminum. So i use 0.25mm CCA wire to wind 150turn each.
I connect the foil to anode 600V and warp with kapton tape for insulation.
First picture show on screen. I have modified the Vertical amplifier by change the Emitter resistor 4.7K to 2.7K to improve the vertical sweep output . The picture is slightly off to right.
I cut a small piece of flexible magnet to place on the tube neck to shift the picture to left to make whole picture stay at center of the screen. I also install speaker.
I increase the vertical amplifier gain by change Re to 2K , the screen is too small and my camera have difficulty to focus on green object in reality the picture is better. but the picture is cant compare with normal TV since 2AP1 are not design for TV purpose.
I try make another video , try my best to get as sharp as possible..
The whole PCB is very small. i like the PCB can put on my hand. The flyback transformer after soap with varnish it become more quiet.
GeeSeang's Hobby site 