Remote repeater - bit more power to led output

[Bob Latham]

Hi,

I'd be grateful for a little advice.

I use a tvLINKplus device for repeating the magic eye remote signals
which are fed back through the UHF aerial cable to my satellite box.

It works OK but the output LEDs have to be fitted in exactly the
right place on whatever kit you're using or it doesn't work and even
then, it's not as reliable as it could be. I Would also like to drive
3 leds for 3 devices.

So I think I could do with a bit more drive current.

I don't have the diagram of the tvLINKplus but by placing resistive
loads on the output it looks like there is the equivalent of a 2,200
ohm resistor in series with the output.

I get a series of pulses which modulate a 38Khz carrier.

The pulses
with no load are only +350mV
When loaded with the original 2 leds this drops to 77mV

I was thinking of using an emitter follower circuit and connecting
the base of the NPN transistor to the output of the tvLINK. I then
realised that the transistor will need 0.6v base-emitter to switch it
on and I've only got 350mV.

I need a simple circuit to give me a bit more drive, 77mV at 0.12mA
is very small indeed to drive leds I was hoping to double it and make
it 3 leds.

Any advice much appreciated.

Thanks.

Bob.

--
Bob Latham
Stourbridge, West Midlands

 

[Jasen Betts]

On 2020-01-15, Bob Latham <bob@sick-of-spam.invalid> wrote:

Hi,

I'd be grateful for a little advice.

I use a tvLINKplus device for repeating the magic eye remote signals
which are fed back through the UHF aerial cable to my satellite box.

I found this picture, is that your setup?
http://www.tvlink.co.uk/downloads/185041484_tvlp.pdf

It works OK but the output LEDs have to be fitted in exactly the
right place on whatever kit you're using or it doesn't work and even
then, it's not as reliable as it could be. I Would also like to drive
3 leds for 3 devices.

So I think I could do with a bit more drive current.

I don't have the diagram of the tvLINKplus but by placing resistive
loads on the output it looks like there is the equivalent of a 2,200
ohm resistor in series with the output.

I get a series of pulses which modulate a 38Khz carrier.

The pulses
with no load are only +350mV
When loaded with the original 2 leds this drops to 77mV

I am of the opinion that an infrared LED won't get out of bed for less
than 1000mV - how are you making these measurements?

Are you seeing DC voltage on the coax going into the TVLINK that connects to
the eye?

It should be possible to drive a silicon transistor in common emitter mode
from the LED drive and use that to drive several LEDs (or to drive a
single LED that floods the whole room, so long as there is no leakage
path back to the remote end)

it might be easy to tap into the DC supply to the tvLINKplus to
run that, or it might be difficult.


--
Jasen.

 

[Bob Latham]

In article <qvooa1$kri$1@gonzo.revmaps.no-ip.org>,
Jasen Betts <jasen@xnet.co.nz> wrote:

On 2020-01-15, Bob Latham <bob@sick-of-spam.invalid> wrote:
Hi,

I'd be grateful for a little advice.

I use a tvLINKplus device for repeating the magic eye remote
signals which are fed back through the UHF aerial cable to my
satellite box.

I found this picture, is that your setup?
http://www.tvlink.co.uk/downloads/185041484_tvlp.pdf

Yes, indeed it is.

It works OK but the output LEDs have to be fitted in exactly the
right place on whatever kit you're using or it doesn't work and
even then, it's not as reliable as it could be. I Would also like
to drive 3 leds for 3 devices.

So I think I could do with a bit more drive current.

I don't have the diagram of the tvLINKplus but by placing
resistive loads on the output it looks like there is the
equivalent of a 2,200 ohm resistor in series with the output.

I get a series of pulses which modulate a 38Khz carrier.

The pulses with no load are only +350mV When loaded with the
original 2 leds this drops to 77mV

I am of the opinion that an infrared LED won't get out of bed for
less than 1000mV - how are you making these measurements?

Before yesterday I would have agreed with you 100%. LEDs to me are a
couple of volts at 10-20mA.

To do the measurements I used a scope and a digital (fluke) meter. I
realised that the meter would not read high enough because it was
pulses and not static but I considered the information useful as a
confirmation. The real measurements were done with a scope turned to
"calibrate". Happy to accept my measurements are 10-20% error
possible but I can't see how it could be an order of magnitude.

First off, the scope said 77mV peak when driving the leds on what
looks like a square wave but of course there is 38khz hiding in there.
With no load the square wave goes to 350mV. I then set out to
calculate the "effective" resistance of the internal output limiter
resistor. I did this by finding the value that dropped the output
voltage by 50% and this was 2200 ohms. The only 'sensible' expected
reading of the day. The fluke was showing about 80% voltage of what
the scope was saying for all readings.

Are you seeing DC voltage on the coax going into the TVLINK that
connects to the eye?

Yes, both in the mock up, which is exactly as the pdf you supplied
and in the actual which has an 8 way distribution amp with remote
pass back which sits between the tvLINK and the tvs. There is around
9 or 10 volts everywhere you would expect. The magic eyes are
actually powered from the distribution amp so the DC leaving the
tvLINK is not used in the real situation.

It should be possible to drive a silicon transistor in common
emitter mode from the LED drive and use that to drive several LEDs
(or to drive a single LED that floods the whole room, so long as
there is no leakage path back to the remote end)

Love the sound of flooding the room.

I'm sure you're correct but I can't work out how to do it. I'm stuck
on the 0.6v base emitter voltage to get the thing to turn on, where
is that going to come from?

I was naively expecting to find 10v pulses of 38khz which would be
passed through a current limiting resistor to feed the LEDs. So I was
thinking emitter follower to up the current so I could set a higher
current with my own limiting resistor. Not enough output voltage.

If I amplify the voltage, surely I then have to go AC, bias the
transistor myself and then couple it AC to the tvLINK via a capacitor
to I can shift the DC bias on the base? I worry that ac coupling will
soften the edges of the square wave.

I admit I'm now out of my depth, hence my plea for help.

it might be easy to tap into the DC supply to the tvLINKplus to
run that, or it might be difficult.

I can get in the tvLINK, the case pops open but the board is very
tight and all surface mount, lots for your money, chips transistors
and no diagram. I'm sure I can get to the power though but still
stumped getting more output.

BTW, much appreciate your time and help, thank you.


Cheers,

Bob.

--
Bob Latham
Stourbridge, West Midlands

 

[Bob Latham]

In article <583337bbc4bob@sick-of-spam.invalid>,
Bob Latham <bob@sick-of-spam.invalid> wrote:

I am of the opinion that an infrared LED won't get out of bed for
less than 1000mV - how are you making these measurements?

Wrongly.

After reading your post this morning I started to think the results I
saw needed checking.

Another nagging worry was the fluke showing roughly 80% of what the
scope said. If it was a square wave it should should 50% but it was
bursts of square waves so should be much less than 50%.

I checked the scope,
set to calibrate - yes.
x10 turned off - yes.

Noticed the scope was set to 50mV /div. So I got a 1.5v battery and
connected it to the probe expecting the display to vanish off the
top. It didn't, it went up 2.7 divisions.

What!

Looked at the probe, the only thing on it is:
Tektronix PR19C TKG 88/MK3

I tried another probe, and you'll not be surprised to learn it read
10 times higher.

So yes, didn't know that probe divided by 10, in my defence I
purchased from a former employer in 1993, I think I've used it twice
since. The probe does not indicate it divides by 10.

So, it looks like I have 3.5v of square wave output with no load
which I could feed into an emitter follower to up the current
available.

Talking of current, all my measurements yesterday were done by seeing
the voltage across known resistors WITH THE SAME PROBE so they are
all 10:1 out as well.

With that info, do you think I could get away with using say a BC108
as emiiter follower, base connected to tvLINK output, collector to
+ve supply emitter to limit resistor and then leds.

I believe the two leds I have at the moment are wired in series, vert
hard to prove without destroying the led cable. Can I get 3 leds to
work with a bit more power?

Any thoughts now you have more reasonable information to go on?

Once again, thanks for your help.


Thanks.

Cheers,

Bob.

--
Bob Latham
Stourbridge, West Midlands

 

[Jasen Betts]

On 2020-01-16, Bob Latham <bob@sick-of-spam.invalid> wrote:

In article <583337bbc4bob@sick-of-spam.invalid>,
Bob Latham <bob@sick-of-spam.invalid> wrote:

I am of the opinion that an infrared LED won't get out of bed for
less than 1000mV - how are you making these measurements?

Wrongly.

Shit happens. good that you sorted it out.

With that info, do you think I could get away with using say a BC108
as emiiter follower, base connected to tvLINK output, collector to
+ve supply emitter to limit resistor and then leds.

BC108 is an antique, but yeah it should work.

emitter follower should work, but common emitter could be worth
experimenting with too - you get more current into the base so
potentially more gain. but 20mA into an LED will probably flood the
whole rack if you aim it at the back wall or the ceiling.

--
Jasen.

 

[Bob Latham]

In article <qvri95$4l6$1@gonzo.revmaps.no-ip.org>,
Jasen Betts <jasen@xnet.co.nz> wrote:

On 2020-01-16, Bob Latham <bob@sick-of-spam.invalid> wrote:
In article <583337bbc4bob@sick-of-spam.invalid>,
Bob Latham <bob@sick-of-spam.invalid> wrote:

I am of the opinion that an infrared LED won't get out of bed for
less than 1000mV - how are you making these measurements?

Wrongly.

Shit happens. good that you sorted it out.

Generous of you - thanks.

With that info, do you think I could get away with using say a
BC108 as emiiter follower, base connected to tvLINK output,
collector to +ve supply emitter to limit resistor and then leds.

BC108 is an antique, but yeah it should work.

Indeed but I might have one in my tat box and I'm an antique.

Do you have a better device in mind?

emitter follower should work, but common emitter could be worth
experimenting with too - you get more current into the base so
potentially more gain. but 20mA into an LED will probably flood
the whole rack if you aim it at the back wall or the ceiling.

Good to hear, thanks for your help.

Cheers,

Bob.

--
Bob Latham
Stourbridge, West Midlands

 

[Jasen Betts]

On 2020-01-17, Bob Latham <bob@sick-of-spam.invalid> wrote:

In article <qvri95$4l6$1@gonzo.revmaps.no-ip.org>,
Jasen Betts <jasen@xnet.co.nz> wrote:
On 2020-01-16, Bob Latham <bob@sick-of-spam.invalid> wrote:
In article <583337bbc4bob@sick-of-spam.invalid>,
Bob Latham <bob@sick-of-spam.invalid> wrote:

I am of the opinion that an infrared LED won't get out of bed for
less than 1000mV - how are you making these measurements?

Wrongly.

Shit happens. good that you sorted it out.

Generous of you - thanks.

With that info, do you think I could get away with using say a
BC108 as emiiter follower, base connected to tvLINK output,
collector to +ve supply emitter to limit resistor and then leds.

BC108 is an antique, but yeah it should work.

Indeed but I might have one in my tat box and I'm an antique.

Do you have a better device in mind?

BC108 is easily sufficient. 2N3904 and BC547C are cheap, but not chaper than
"I have one in my tat box"

emitter follower should work, but common emitter could be worth
experimenting with too - you get more current into the base so
potentially more gain. but 20mA into an LED will probably flood
the whole rack if you aim it at the back wall or the ceiling.

--
Jasen.

 

[Bob Latham]

In article <qvsvb8$kc9$1@gonzo.revmaps.no-ip.org>,
Jasen Betts <jasen@xnet.co.nz> wrote:

On 2020-01-17, Bob Latham <bob@sick-of-spam.invalid> wrote:

Do you have a better device in mind?

BC108 is easily sufficient. 2N3904 and BC547C are cheap,

Thanks for those, I'll look them up for spec and price.

> but not chaper than "I have one in my tat box"

Indeed.

Thanks.

Bob.

--
Bob Latham
Stourbridge, West Midlands