Photomultiplier Tubes camera build

[Harry Dalek]

Today i changed the 22k resistor at the emitter of the opto reflector for the 47k trimmer on Klaas suggestion the did help i could see on the scope adjusting the signal to correct waveform shape amplitude .
So Thanks Gary Klass getting better ,it got to the point to day where there was little movement when i switched to crystal sync to the video sync mixer...
I think the motor control side still needs work so i don't have to adjust each time .but its much better today any way .
So i took the opportunity and did some tests just as the encoder was in sync with the crystal as far as it being locked i will see next time i switch it on so thats debatable...
The Pictures are a touch off half way down as i stuffed the encoder black and white position again perhaps one more go and it will be right !
Lighting was a touch better today does not take much for it to go either way even putting a white sheet of paper in front of the camera is enough at times to overload it .
This was again just room lighting from a window ...
I inverted every thing so it should play fine ...
first wav is taken at the out put of pin 7 lm311 again this time after Klaas advice .
Then a bit of fun with less movement as i could see you need to move slow for nbtv .
All the video wav files i had the sync on crystal apart from 2nd gary klass wav

[gary]

Harry it's a real pity that you didn't have an encoder version when the disk was steady - also how did you stop the slow roll? using the 22pf trimmer?

It would be good if you switched between crystal and encoder syncs half way like you did before.

It seems that the sync bar is halfway whether it's crystal or encoder so I would think that means it a detector/encoder alignment issue.

[Harry Dalek]

Harry it's a real pity that you didn't have an encoder version when the disk was steady - also how did you stop the slow roll? using the 22pf trimmer?

Well i did try adjusting it before but it could just be dumb luck !....

What i do to check the disk speed i use the scope sound card scope program and go to the frequency bandwidth screen ..i have the probe on the sync mixer ...i can see the disks speed in frequency and its hunting and i can see how its matching to the crystal clock this gos to a non moving spike.
The 2 spikes match each other at 400hz its like putting one finger in front of the other and moving one the one not moving is the crystal the other the encoder swaying when that stops and both are dead over each other which is pretty hard and has not happened often i know there will be no movement when i use the crystal as sync ..

Pretty much to day i swapped that resistor for a trimmer set it so the out put of the opto reflector was a pure sinewave of highest amplitude then checked the output of the lm311 nice square waves looked pretty even to the eye as far as pulse width and seeing missing pulse then adjusted speed as best i could and looks like at that stage the settings were right spot on ...but i have to see when i switch on again it does the same if it gos a touch random again i know i still need to fixed some thing in the motor speed area i suppose ,

It would be good if you switched between crystal and encoder syncs half way like you did before.

I will do that again for ya ..have to go out tomorrow but when i have time before or after i will give it a go .

It seems that the sync bar is halfway whether it's crystal or encoder so I would think that means it a detector/encoder alignment issue.

I have been thinking about this and i thought i put the reflective white to the start of the line again if thats correct but i wonder if theres over lap the encoder is to large and other lines are being over lapped ...as i said curse of the encoder it will never end till i make every mistake

[gary]

I have been thinking about this and i thought i put the reflective white to the start of the line again if thats correct but i wonder if theres over lap the encoder is to large and other lines are being over lapped ...as i said curse of the encoder it will never end till i make every mistake

Is there any chance of a photo showing the detector and encoder front on (or rear on depending on how you have it)? Or is that impossible without pulling things apart?

[Harry Dalek]

Is there any chance of a photo showing the detector and encoder front on (or rear on depending on how you have it)? Or is that impossible without pulling things apart?

I took a photo today it was pulled apart as i had to get the PMT hole larger so i can adjust its position and stop some more light its back together again for testing ...i did try again with a slightly smaller encoder so the black and white fit between the lines start and end closer ..i have only given it one run today this after noon..

I will see what i can do on the photos

Have a see how i go with this one if its still off ...i did take a photo of it ...

What i did also today was matt black paint the pmt tubes clear glass to stop any light getting in here...

EDit
Test run to day to me the shading is better but i still have the tunnel look so it must be the other side of Nipkow lens side ..may be if i move the nipkow towards the lens it might help //
Still also have the picture half out of wack and it still hard to keep on 400 hz so more work on all this week ....reason also Gary i didn't do the crystal encoder test to night ....Back to encoder number 5 or 6 by now

[Klaas Robers]

Harry, I see that you have a missing sync pulse on your encoder disc. You use this output for the speedcontrol of the disc, however you compare it against an output signal of the 4060, which has NO missing pulse. That will not match and introduces at least speed variations on your disc.

So, unless you manage to make a missing pulse,every 12.5 sec, in the 400 Hz signal out of the 4060 (this is very difficult) the encoder should give uninterrupted sync pulses. For the missing sync pulse you could make an extra ring on the encoder disc, with only one pulse. In an electronic way this one sync pulse per revolution may suppress one sync pulse from the 32 on the outer ring.

[gary]

Harry, I see that you have a missing sync pulse on your encoder disc. You use this output for the speedcontrol of the disc, however you compare it against an output signal of the 4060, which has NO missing pulse. That will not match and introduces at least speed variations on your disc.

So, unless you manage to make a missing pulse,every 12.5 sec, in the 400 Hz signal out of the 4060 (this is very difficult) the encoder should give uninterrupted sync pulses. For the missing sync pulse you could make an extra ring on the encoder disc, with only one pulse. In an electronic way this one sync pulse per revolution may suppress one sync pulse from the 32 on the outer ring.

That's a very good point Klaas, I don't know how I missed that myself.

[Harry Dalek]

Harry, I see that you have a missing sync pulse on your encoder disc. You use this output for the speedcontrol of the disc, however you compare it against an output signal of the 4060, which has NO missing pulse. That will not match and introduces at least speed variations on your disc.

Original one didn't thats used for framing yes ?...thats a problem to frame or not to frame that is the question ~
Klass curse of the encoder once again....it has found another problem ...
I see what you are saying ...out of interest a nbtv video signal has the missing pulse for framing init ...some thing i never thought about ..

oS, unless you manage to make a missing pulse,every 12.5 sec, in the 400 Hz signal out of the 4060 (this is very difficult) the encoder should give uninterrupted sync pulses. For the missing sync pulse you could make an extra ring on the encoder disc, with only one pulse. In an electronic way this one sync pulse per revolution may suppress one sync pulse from the 32 on the outer ring.

[/quote][/quote

I can do that with the 4060 but not with a crystal but a resistor capacitor oscillator ...i think its one of your old circuits in a test pattern generator i have used before on 2 motor system...

Perhaps to make it simple and forget about framing for now .

The main problem when you switch sync to the crystal side is is the motor control via the encoder feed back circuit at 400hz has to be dead on ...if it is there will be no difference apart from better sync pulses ...i see what you are saying my clock is a pulse every time where as the encoder is adding a long pulse to throw it off....motor control ?

I see most people that make cameras have a closed-circuit television i have seen wav examples of camera to video signal so perhaps not a problem for some

[Harry Dalek]

OK i see i need to change the missing pulse back to a pulse as an experiment and see if it helps .

[Harry Dalek]

Well Klass and Gary i tried the an encoder with no missing sync and it was possible to sync the encoder to the crystal...... today any way

So Gary i did that test you wanted the other day the chicken test is more likely the better one.

The encode one does not look to bad apart towards end half of the right side perhaps its seeing the slight change in the glued on pulse or i need to check the encoder is glued flat .

OH well getting there !

Ive been thinking about the tunnel look problem i may look into making another CD hood with a different lens .

[gary]

Hi Harry, how far is the tip of the detector from the encoder?

[Harry Dalek]

Hi Harry, how far is the tip of the detector from the encoder?

i would say about 5mm i tried pulling it away while looking on the scope and i lost signal this was giving best result on the scope i had to angle it a touch as well.....the the sine wave signal looks pretty neat from the opto reflector HOA1405....i checked the data sheet seems thats about right as well./

[gary]

Hi Harry, how far is the tip of the detector from the encoder?

i would say about 5mm i tried pulling it away while looking on the scope and i lost signal this was giving best result on the scope i had to angle it a touch as well.....the the sine wave signal looks pretty neat from the opto reflector HOA1405....i checked the data sheet seems thats about right as well./

Ok, from memory that is the optimum distance.

I am having trouble coming up for a "physical" explanation for the waviness of the sync bar when taken from the encoder. I thought it was an artefact of the "hunting" you were having but in these tests the hunting changes from none or nearly none to fairly severe but the amplitude of the waviness appears to remain the same.

Harry, this is a long shot, could you check, with a ruler, that the encoder is circular rather than elliptical - the picture a few posts back seems to indicate it is quite elliptical but that could be just the angle of the camera - but some printers have been known to print anisotropically. However, it's hard to imagine even that would result in the variation we are seeing.

The wav files do indicate that you have a substantial 'glitch' where you added the extra segment, which probably doesn't help in speed stability either.

I found this that I think I used on one of my cameras:

true black encoder.bmp

(3.61 MiB) Downloaded 180 times

[Harry Dalek]

Ok, from memory that is the optimum distance.

I was thinking i might been off but i am recalling the first go where i was way off on the flying spot camera go you did mention in a post there 5mm was more correct than what i was doing at the time ...

I am having trouble coming up for a "physical" explanation for the waviness of the sync bar when taken from the encoder. I thought it was an artefact of the "hunting" you were having but in these tests the hunting changes from none or nearly none to fairly severe but the amplitude of the waviness appears to remain the same.

When i stick the encoder on I try and line the encoder up to center so theres not a difference in pulse width as the encoder rotates as i think would happen if its been put on off center .
But having said that Gary i am doing it by eye and i understand close is not good enough if as now it looks like theres another encoder problem ....if theres one i will sure find it

Harry, this is a long shot, could you check, with a ruler, that the encoder is circular rather than elliptical - the picture a few posts back seems to indicate it is quite elliptical but that could be just the angle of the camera - but some printers have been known to print anisotropically. However, it's hard to imagine even that would result in the variation we are seeing.

yes i can check that i will also check the distance of the end of each pulse the black as i can see where that ends on the encoder paper to the end of the cd ..that way we can tell if theres a pulse width change as the disk rotates ...should be the same for every pulse .

The wav files do indicate that you have a substantial 'glitch' where you added the extra segment, which probably doesn't help in speed stability either.

Yes for sure another by eye stick on ...Oh well i will know now not to try that idea again !

I found this that I think I used on one of my cameras:

[/quote][/quote]

Thanks for that Gary ! i will print it out and try it but i will for science do the readings on this encoder

[Harry Dalek]

OK the encoder i printed is 8.7cm cross to the end of the pulse other side of the disk looked at a number all the same so the encoder is printed round...

Now is the encoder centered correct lets take a reading of the 32 pulses..slight wobble happening i can see ..readings i ended up taking from the bottom of the black pulse to the edge of the cd that seemed easiest
1...2.5cm
2...2.5cm
3...2.4cm
4...2,4cm
5---2.4cm
6---2.4cm
7---2.4cm
8----2.4cm
9 ...2.4cm
10...2.4cm
11...2.4cm
12,,2,4cm
13...2.4cm
14...2.4and half a mm
15....2.5cm
16....2.5cm
17....2.5cm
18...2.5cm
19...2.5cm
20...2.5cm
21...2.5cm
22...2.5cm
23my stick on pulse 2.5cm
24...2.5cm
25...2,5cm
26...2.5cm
27...2.5cm
28...2,5cm
29...2.5cm
30...2.5cm
31...2.5cm
32...2.5cm