Starting at the beginning with Mirror Screws

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Dave Moll, Andrew Davie, Steve Anderson

Re: Starting at the beginning with Mirror Screws

Postby Steve Anderson » Wed Sep 09, 2020 3:47 pm

Panrock wrote:...Your circuit remains sharp and loafs along...

I don't recall what the bandwidth of those drivers are, but probably OK up to a frame rate of 25Hz, perhaps more...I think it was a function of the cable length between the driver and the LEDs/lasers. This may be something to consider...I guess you'd prefer the controls for gain and black level remote from the display, correct?

These ones using the BD1xx's will be limited by the PWM arrangement, somewhere in the region of 200kHz...but note the cable length remark above. Which reminds me - I must look into taking the PWM to 500kHz, it's slipped my mind over the past few days.

Later...It hasn't slipped my mind, the parts required are already on my shopping list...task multiplexing...something guys are generally bad at!

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Starting at the beginning with Mirror Screws

Postby Steve Anderson » Wed Sep 09, 2020 4:44 pm

Hmmm, this cable length issue has me a little concerned. It may be that we put the comparators, ramp generator and the drivers inside the display, have the gain/black controls remote and link the two via three matched 75R coax cables, plus power if required. No rocket science, it just needs looking into...RG179 is the cable I would suggest, but RS (and probably others) only sell it by the reel of 100m, which is expensive. It does need to be video cable, not RG59 RF cable. The difference is the amount of braiding/screening coverage, plus usually a different type of dielectric.

Or, we change it to a RS485 differential driven cable arrangement and use one CAT-5 or 6 cable for all three signals...that may be a better way to go...in fact thinking it over for the past 10 minutes, it is the way I would go...suggested RS485 driver/receivers would be MAX485, (surprisingly) you'd be needing six of them, RS 190-1041, plus 6 110R resistors. They're good up to 2.5Mbs. I'm looking at other devices, so don't go placing orders yet..

OK, I suggest sticking with the MAX485s, they're a known quantity to me, but hold off ordering for a little while...

Your thoughts?

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Starting at the beginning with Mirror Screws

Postby Panrock » Wed Sep 09, 2020 7:25 pm

Thanks for this. Basically it's undesirable to have any electronics at the screw, for physical safety reasons if servicing while it's in operation. This whizzing 20kg stainless steel slicing machine could easily take your hand off - no sweat.

Having said that, the distance between the screw and the new screen will be about far less than the old viewing distance - was about 5 metres away - screen now 2.3 metres distant. Come to think of it, all this is irrelevant because the laser diodes will be pointing toward the screw from a position under the screen, and their driving electronics and controls could be directly adjacent. So, no distance issues whatsoever.

Steve O
Panrock
Green padded cells are quite homely.
 
Posts: 870
Joined: Mon Feb 05, 2007 8:25 am
Location: Sedgeberrow, England

Re: Starting at the beginning with Mirror Screws

Postby Steve Anderson » Wed Sep 09, 2020 7:44 pm

Err...are you saying that all controls will be on the display unit itself? No remote unit as previous? It's no problem either way, but let me get it correct. It would be like going back to the days before remotes, you had to get out of your armchair to change channels or adjust the volume...

If you could do a rough sketch (plan view) where things (and people) are placed in relation to each other, that would help me in understanding what your concept is.

Unfortunately you'll have to have some electronics near the screw, within 30cm of the laser diodes, but can you not disconnect the motor if the electronics needs some attention? There'll be no need for pre-sets or other adjustments within the display unit. And/or have the electronics as a removable board or module?

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Starting at the beginning with Mirror Screws

Postby Panrock » Wed Sep 09, 2020 10:51 pm

Hopefully, this will make things clearer.

The only thing needed near the screw will be the (existing) position sensor, needed for sync.

Steve O
Attachments
Mirror Screw.gif
Mirror Screw.gif (18.7 KiB) Viewed 8600 times
Panrock
Green padded cells are quite homely.
 
Posts: 870
Joined: Mon Feb 05, 2007 8:25 am
Location: Sedgeberrow, England

Re: Starting at the beginning with Mirror Screws

Postby Steve Anderson » Thu Sep 10, 2020 12:09 am

Ah! I think I comprehend now...so the only electrical items within the 'box' that contains the mirror screw is the motor and the sync/position sensor? All the gubbins I'm dealing with are at a distance, including the laser diodes? Correct? If so that makes things a lot simpler, as long as any optical hiccups can be overcome. Though that's not not my expertise...

As it is said, "A picture is worth..." and so on. So no long trailing leads, no electronics in the mirror-screw box beyond those mentioned already. It's starting to set in my gelatinous mind.

In the last few hours I've changed some details around such that the BD1xx's never go into saturation as previously which eliminates/reduces storage and rise-time. They may have to dissipate a bit more heat but they're well within their ratings. The +12VB supply can become unregulated and simply called +18V, the +12VA supply regulator can be fed from this too. The heat which would have been dissipated by the +12VB regulator is now shared across the three BD1xx's.

As ever design is iterative, you eventually get there...especially when there's two of you working around 10,000km apart and at (currently) a six-hour time difference...creepy, but I'm from your future...depressing really...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Starting at the beginning with Mirror Screws

Postby Panrock » Thu Sep 10, 2020 12:19 am

Steve Anderson wrote:creepy, but I'm from your future...depressing really...

Steve A.

As ever, good to work with you. When you get to my age, it's nice to find out you actually have a future... :lol:

Steve O
Panrock
Green padded cells are quite homely.
 
Posts: 870
Joined: Mon Feb 05, 2007 8:25 am
Location: Sedgeberrow, England

Re: Starting at the beginning with Mirror Screws

Postby Steve Anderson » Thu Sep 10, 2020 1:58 pm

This is proving a bit tricky, it's not impossible but it may require a change in how the PWM is done. There are quite a few constraints imposed when you go from 100kHz to 500kHz PWM.

Also a slow-rise power supply for the laser diodes rather than it just 'bangs on'. (LM317),

Plus comparators with an open collector output with enough speed and differential input voltage spec are few and far between.

Given a few days I should be able to resolve these issues...

Steve A.

Steve, see if you can find 74HC541s in DIL/DIP packaging in the UK, RS and Farnell/Element 14 don't seem to have them, though I can get them here...

The open-collector requirement for the comparator outputs may now have gone away...and I may be drifting back to using MOSFET drivers...the very familiar IRF540 as used in many NBTV motor controllers...possibly TO92 SCRs/Thyristors too,,,

As I mentioned before, don't order anything yet...
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Starting at the beginning with Mirror Screws

Postby Panrock » Thu Sep 10, 2020 10:18 pm

No rush Steve, take your time.

I've also had a look at Mouser Electronics but it isn't clear which if any of these might be DIL/DIP.

All the best,

Steve O
Panrock
Green padded cells are quite homely.
 
Posts: 870
Joined: Mon Feb 05, 2007 8:25 am
Location: Sedgeberrow, England

Re: Starting at the beginning with Mirror Screws

Postby Steve Anderson » Fri Sep 11, 2020 12:55 pm

The DIL/DIP package variant is the third one down that Mouser page, part number SN74HC541N, Mouser number 595-SN74HC541N. Texas Instruments indicate a DIL/DIP package by an 'N' as the last character of the part number, other suppliers have their own schemes, as always - different!

Don't order yet, that requirement may have gone away...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Starting at the beginning with Mirror Screws

Postby Steve Anderson » Fri Sep 11, 2020 4:30 pm

The more I get into this the more I'm thinking of an analogue solution rather than using PWM. The vote hasn't been cast yet, but apart from dealing with the 'threshold current' of the laser diodes it should be somewhat simpler...I need to dwell on this some more...

Looking at the datasheets, once the laser diode is above its 'threshold current' it is (they are) linear enough...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Starting at the beginning with Mirror Screws

Postby Panrock » Fri Sep 11, 2020 7:41 pm

Well, this is the same conclusion you reached before, when you designed the - excellent - analogue drivers for the LED version of my mirror screw... also 240 KHz bandwidth of course.

Another question I am unsure of relates to eye safety. E.g. is a repetitive light pulse from a laser going full whack with a mark-space ratio of 1:1000 really as safe as a laser operating continuously at 1/1000 of full brightness? This is not something we want to get wrong.

Steve O
Panrock
Green padded cells are quite homely.
 
Posts: 870
Joined: Mon Feb 05, 2007 8:25 am
Location: Sedgeberrow, England

Re: Starting at the beginning with Mirror Screws

Postby Steve Anderson » Fri Sep 11, 2020 7:52 pm

That I'm no expert in. I would have assumed it's the average dosage, but an human-optical specialist may be qualified to answer that. I most certainly am not! If you wear glasses (I do) a quick trip to your optician for a check-up may be in order. If they don't know the answer they may be able to point you in the right direction.

A friend of mine had his retina in one eye re-attached using lasers a couple of years ago. A specialist eye-hospital may be a source of information.

My local one in Bangkok...and the one he used...

https://www.rutnin.com/en/home/

I would think that the Health & Safety Executive in the UK would have guidelines too...God bless 'em...(I'm being cynical)...you can't even fart without them complaining...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Starting at the beginning with Mirror Screws

Postby Panrock » Fri Sep 11, 2020 9:19 pm

The answer probably has a lot to do with the repeating pulse length, and its relationship to the response time constants of the retina rods and cones. In this case. the repetition rate will be much shorter than the TC of 1/10 second or so - it has to be to obtain the 240KHz frequency response - and we'll be OK.

However, given I already have a control box providing three excellent analogue outputs for the LEDs, what about a "simple" conversion circuit (x3) to make these outputs suitable for feeding laser diodes? This might save both of us work...

Steve O
Panrock
Green padded cells are quite homely.
 
Posts: 870
Joined: Mon Feb 05, 2007 8:25 am
Location: Sedgeberrow, England

Re: Starting at the beginning with Mirror Screws

Postby Steve Anderson » Sat Sep 12, 2020 12:40 pm

Panrock wrote:...what about a "simple" conversion circuit (x3) to make these outputs suitable for feeding laser diodes? This might save both of us work...Steve O

That's a possibility, I'll have to review what was done a few years back, mainly in terms of the bandwidth...

Steve A.

Is the circuit you finally used the one posted in this thread on 28th Nov. 2013? The circuit itself is dated 21st Dec. 2012...it doesn't seem possible that it's almost eight years ago! I thought maybe three or four years ago - not almost eight!

I've looked over the previous postings related to the circuit mentioned above, going by the waveforms posted around the same time the existing bandwidth appears to be in the order of 700kHz, more than I thought and adequate for what's required here.

An approximation for bandwidth is:-

Bandwidth x Rise or Fall time = 0.45

I averaged out the rise/fall times to be 650ns from the waveforms, 0.45/650 x 10^-9 = 692kHz...call it 700kHz, it is an approximation after all.

I'll work on 'massaging' the circuit to suit the laser diodes.

Though I'm a little confuddled whether it was this analouge driver that was used or the PWM method. I must improve my record-keeping!

I'll be able to measure the bandwidth a bit better as soon as I get the AD9850 oscillator operational.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

PreviousNext

Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 7 guests

cron