Searching \ for '[EE] High voltage opto-coupler' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/power.htm?key=voltage
Search entire site for: 'High voltage opto-coupler'.

Exact match. Not showing close matches.
PICList Thread
'[EE] High voltage opto-coupler'
2009\07\05@202033 by Jinx

face picon face
In Figure 3 here

http://www.rory.co.nz/projects/lighting/beat_strobe.html

the author specifies an H11D1/H11D2 optocoupler, which has a
Vce of 300V, as described in "Opto-coupler Selection"

None of my usual suppliers have the H11D1 or 2, and AFAICT
none of their other opto-transistors have Vce greater than 70V. It
isn't possible to do a quick search because output voltage isn't one
of the parameters, only isolation voltage and output current. Only
by downloading all the datasheets

Is there a reason why a MOC3020 opto-triac couldn't be used ?

Or is there an alternative opto > 80V ?

How about a Darlington with a lower voltage opto (eg 4N25 or 4N33
which I have a few of) and an external high voltage transistor ? I think
I would need a voltage divider to knock down the voltage across the
opto's internal transistor. If so, how would this affect the drive for the
SCR gate through the external transistor ?

TIA

2009\07\05@205836 by Richard Prosser

picon face
Hi Jinx,

I would think a std. opto driving a high voltage transistor would be fine.

He says the optos are available at sicom, so if you get stuck I can
shoot down there & get you a couple. (I think they do mail order also
- <http://www.sicom.co.nz>)

Richard



2009/7/6 Jinx <spam_OUTjoecolquittTakeThisOuTspamclear.net.nz>:
{Quote hidden}

> -

2009\07\05@212934 by Jinx

face picon face
> I would think a std. opto driving a high voltage transistor would
> be fine

Thanks, I'm still weighing up options

Trawling through datasheets I think the PC851 might be an
equivalent

http://datasheet.digchip.com/424/424-3-108466-PC851.pdf

Radiospares have them

2009\07\05@235741 by SME

picon face
> Is there a reason why a MOC3020 opto-triac couldn't be used ?
>
> Or is there an alternative opto > 80V ?

Even a 300V opto would be marginal there.

However, you should have no problem tapping down the high voltage with
a resistive divider.
Issues are that gate current can be quite high so that a divider that
is always on could dissipate excessive energy. Adding an extra
transistor or two to translate from opto circuit to gate drive would
allow good drive with low average power dissipation. I can supply an
?MPSA43? which would work there if you want one. (MPSA42 is common.
MPSA4* is it's little seen cousin with an  extra 100 volts rating. (*
= whatever it is I've got - I think it's MPSA43).



        Russell

2009\07\06@022253 by Jinx

face picon face
part 1 1343 bytes content-type:text/plain; (decoded 7bit)

>> Is there a reason why a MOC3020 opto-triac couldn't be used ?

I've come across the attached in an old folder. No idea where it came
from though. Suspect it would be from one of those 'circuit collection'
sites. Comments ?

> Even a 300V opto would be marginal there.
>
> However, you should have no problem tapping down the high voltage
> with a resistive divider.
> Issues are that gate current can be quite high so that a divider that
> is always on could dissipate excessive energy

Other optos I've looked at besides the H11D1 do have 300V Vce
but not its 100mA Ic. If the author is correct that 76V is the highest
expected across the opto then perhaps the 4N38 (80V), which is
covered by the same datasheet as the H11D1 Notwithstanding your
comment about a 300V opto being marginal. If Vc is 70V, why
would a 300V component be at risk ?

> Adding an extra transistor or two to translate from opto circuit to
> gate drive would allow good drive with low average power dissipation.
> I can supply an ?MPSA43? which would work there if you want one.
> (MPSA42 is common. MPSA4* is it's little seen cousin with an
> extra 100 volts rating. (* = whatever it is I've got - I think it's
MPSA43)

You're suggesting an MPSA in a Darlington arrangement with the opto ?


part 2 3869 bytes content-type:image/gif; (decode)


part 3 35 bytes content-type:text/plain; charset="us-ascii"
(decoded 7bit)

2009\07\06@093837 by Spehro Pefhany

picon face
At 08:19 PM 05/07/2009, you wrote:
>In Figure 3 here
>
>http://www.rory.co.nz/projects/lighting/beat_strobe.html
>
>the author specifies an H11D1/H11D2 optocoupler, which has a
>Vce of 300V, as described in "Opto-coupler Selection"
>
>None of my usual suppliers have the H11D1 or 2, and AFAICT
>none of their other opto-transistors have Vce greater than 70V.

A quick parametric search at Digikey turns up more than SIXTY part number of
300/350/400V output transistor output opto-isolators, all in stock.

>Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
.....speffKILLspamspam@spam@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com



2009\07\06@122923 by Dwayne Reid

flavicon
face
I used to build those things in the dark, distant past.  I used
MOC3010 opto-triac back then (now-days I would use a MOC3011) in
place of the original SCR.  Worked extremely well.

The majority of the strobes that I used to modify used a pot wired up
as a voltage divider feeding a RC time delay which then triggered a
diac which in turn fired the SCR.  The reason I mention this is
because I found it useful to NOT remove the original SCR - just leave
the pot set all the way OFF.  That disabled the original
self-contained flashing and allowed complete remote control of the
strobe via the opto.  The triac in the opto was simply connected in
parallel with the original SCR.

The advantage of doing it that way was that the strobe could still be
used for its original purpose.

I still have a bunch of those strobes around - I now use them for
bright call indicators for Intercom systems.

The reason I modified strobes that I purchased instead of building
from scratch were several: existing certification on the purchased
strobe meant that getting the modified strobe certified by the local
inspection authority was relatively cheap and, most important, the
purchased strobe cost far less than the parts cost and labor for me
to build from scratch.  The last time I purchased these, I think that
they cost something like Can $10.00 each at the local grocery store
(they were being sold just before Halloween).  They even came with 3
different colored gels that could be slid over the front.

Hope this helps!

dwayne


At 06:19 PM 7/5/2009, Jinx wrote:

{Quote hidden}

--
Dwayne Reid   <dwaynerspamKILLspamplanet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(780) 489-3199 voice          (780) 487-6397 fax
http://www.trinity-electronics.com
Custom Electronics Design and Manufacturing

2009\07\06@203022 by Jinx

face picon face
> A quick parametric search at Digikey turns up more than SIXTY
> part number of 300/350/400V output transistor output opto-isolators,
> all in stock

Spehro, Digikey have representation in New Zealand, as do many other
companies like them. Looking at the search results for H11D1 they have
many variants, and at reasonable prices. The problem is that it appears
MOQ is a tube. For 6-pin DIP that's quite a few. If Digikey is like Farnell
for example, they have no physical presence in NZ to trot along to for
ones and twos

I've decided to not mess about with making a HV opto and just follow
the original circuit. Active Components here can get me 1-up at an OK
price ($1.50, approx Digikey bulk price) so I'll add 10 to an order

One implementation I'm thinking of is a bank of strobes triggered in
a circular sequence. As each strobe's capacitor has more time to
accumulate charge this should be a way of increasing flash rate whilst
retaining brightness

2009\07\06@203022 by Jinx

face picon face
> I used to build those things in the dark, distant past.  I used
> MOC3010 opto-triac back then in place of the original SCR.
> Worked extremely well

Dwayne, I built a stage strobe a looooong time ago (79 ?), from a
design in ETI. It used the traditional method of neon breakdown to
trigger the SCR via an ST2 diac. ISTR it used a BT131 as the
dump switch for 2uF. I'm not sure how far back the MOC3010
goes but I wonder why something as relatively hefty as a BT131
would be used. Although many-amp SCRs aren't that expensive
so it's cheap over-engineering



2009\07\16@225016 by Jinx

face picon face
Picked up my H11D1 optos for Figure 3 at

http://www.rory.co.nz/projects/lighting/beat_strobe.html

and been trying, unsuccessfully so far, to get a flash out of it

The problem appears to be that the SCR gate rises to only 200mV
when the opto is activated. Looking on the scope there don't seem
to be any gate pulses higher than 200mV and no activity at all on the
cathode. Voltage on the cold side of the 2k2 (opto collector) also
drops to near-zero

With the transformer out however, the gate sees 65V (V+ = 310V).
Using a 470R resistor as a light load instead of the transformer the
gate voltage again drops to 200mV

Transformer datasheet

http://www.jaycar.co.nz/products_uploaded/MM2520.pdf

It measures OK. Something under 1R on the primary and 17R on
the secondary. I've tried a couple of BT151 800R SCRs, rated at
100A peak. It seems a bit unlikely that a-k is broken, given the
energy available from C15

Any ideas ? The author's pdfs show no extra, omitted or changed
components so of course I wonder why his works and mine doesn't.
My only thought would be a separate 65V supply for the opto/gate
that won't drain C15 (if that's what's happening now)

TIA


2009\07\17@005013 by AGSCalabrese

picon face
#1)   Are you sure it is a 2.2K resistor for R20 ?
#2)   What happens when you short across the output transistor of  
U4 ?  ** Too bad the schematic does not show pin numbers.
                Does Q1 conduct ?  If it does not, then Q1 is bad or  
R20 and D2 and U4 ( transistor ) do not supply sufficient drive to  
trigger Q1.
                If Q1 conducts when shorting the U4 transistor, then  
perhaps U4 is not being driven hard enough.  You could add an external  
transistor to increase the current flow through D2.

** do not use your tongue
Gus


On Jul 16, 2009, at 8:49 PM, Jinx wrote:

Picked up my H11D1 optos for Figure 3 at

http://www.rory.co.nz/projects/lighting/beat_strobe.html

and been trying, unsuccessfully so far, to get a flash out of it

The problem appears to be that the SCR gate rises to only 200mV
when the opto is activated. Looking on the scope there don't seem
to be any gate pulses higher than 200mV and no activity at all on the
cathode. Voltage on the cold side of the 2k2 (opto collector) also
drops to near-zero

With the transformer out however, the gate sees 65V (V+ = 310V).
Using a 470R resistor as a light load instead of the transformer the
gate voltage again drops to 200mV

Transformer datasheet

http://www.jaycar.co.nz/products_uploaded/MM2520.pdf

It measures OK. Something under 1R on the primary and 17R on
the secondary. I've tried a couple of BT151 800R SCRs, rated at
100A peak. It seems a bit unlikely that a-k is broken, given the
energy available from C15

Any ideas ? The author's pdfs show no extra, omitted or changed
components so of course I wonder why his works and mine doesn't.
My only thought would be a separate 65V supply for the opto/gate
that won't drain C15 (if that's what's happening now)

TIA


2009\07\17@005358 by Peter van Hoof

face picon face

I would initially test replacing the thyristor with a momentary switch
from anode to cathode. Be careful!

If you can trigger the tube by activating the switch the t1, r21 and c15 are ok.

If you can not, expect a problem with the capacitor. try a different one. You could
try raising its value to 220 nf or so.

If you are able to trigger put the thyristor back in , pull the optocoupler out.
put the  momentary switch in the spot where the opto's transistor is.

If you can still trigger the circuit the thyristor is ok and the problem must be
with the opto or the quality of the pulse to the led of the opto.

If you can not trigger the flash with the switch the current into the gate may
be insufficient for the thyristor, check diode polarity, double check r20's value and
check the thyristors gate to cathode diode.

Peter


{Original Message removed}

2009\07\17@005546 by AGSCalabrese

picon face
When U4 is activated the voltage across the U4 transistor output  
should drop from 65V to a least below 60V.
Gus


On Jul 16, 2009, at 8:49 PM, Jinx wrote:

Picked up my H11D1 optos for Figure 3 at

http://www.rory.co.nz/projects/lighting/beat_strobe.html

and been trying, unsuccessfully so far, to get a flash out of it

The problem appears to be that the SCR gate rises to only 200mV
when the opto is activated. Looking on the scope there don't seem
to be any gate pulses higher than 200mV and no activity at all on the
cathode. Voltage on the cold side of the 2k2 (opto collector) also
drops to near-zero

With the transformer out however, the gate sees 65V (V+ = 310V).
Using a 470R resistor as a light load instead of the transformer the
gate voltage again drops to 200mV

Transformer datasheet

http://www.jaycar.co.nz/products_uploaded/MM2520.pdf

It measures OK. Something under 1R on the primary and 17R on
the secondary. I've tried a couple of BT151 800R SCRs, rated at
100A peak. It seems a bit unlikely that a-k is broken, given the
energy available from C15

Any ideas ? The author's pdfs show no extra, omitted or changed
components so of course I wonder why his works and mine doesn't.
My only thought would be a separate 65V supply for the opto/gate
that won't drain C15 (if that's what's happening now)

TIA


2009\07\17@035813 by Alan B. Pearce

face picon face
>With the transformer out however, the gate sees 65V (V+ = 310V).
>Using a 470R resistor as a light load instead of the transformer
>the gate voltage again drops to 200mV

This sounds like something in the trigger circuit is always turning on the
SCR. You should be able to verify that this is the case by measuring the
voltage between the 1n4007 and the opto. Also I would expect a resistor of
about 10k between gate and cathode of the scr to prevent false triggering of
the scr by stray emf/emi/induced hum.

I don't understand why he has the 1n4007 in there either. this seems
unnecessary.

2009\07\17@055345 by Jinx

face picon face
> #1)   Are you sure it is a 2.2K resistor for R20 ?

Yes

> #2)   What happens when you short across the output transistor of  
> U4 ?

No difference if the opto is replaced with a pushbutton, or if the SCR
is replaced with a pushbutton. ie supplying C15's charge straight to
the transformer

2009\07\17@055349 by Jinx

face picon face
> I would initially test replacing the thyristor with a momentary switch
> from anode to cathode. Be careful!
>
> If you can trigger the tube by activating the switch the t1, r21 and
> c15 are ok.

Tried that. The SCR was replaced with a pushbutton between the anode
and cathode pads. I see a very very short pulse at the transformer primary.
Not easy to see on the scope, but I reckon about 20V for less than 1us

Won't attempt to measure the secondary output in case something goes
right and HV takes out the scope input. But it's not enough to trigger the
flash anyway

> If you can not, expect a problem with the capacitor. try a different one

Bumped up the 0u1 to 0u16, no change

Tested the SCR. a-k both directions is > 50M, g-a > 50M one way
and 40M the other. Tried 3 optos and 3 SCRs. My guess is at this stage
it's too little current rather than too much blowing components


2009\07\17@055412 by Jinx

face picon face
> When U4 is activated the voltage across the U4 transistor output  
> should drop from 65V to a least below 60V.

Yes, I'd expect that as C15 discharges. I might expect though that
some sort of discharge curve would be visible, not a near-vertical
nose-dive

2009\07\17@071707 by Peter van Hoof

face picon face

Is the trigger transformer close to the bulb?
Any chance of arcing from the trigger wire
Did you try to reverse the two main electrodes of the bulb?
At this point a problem with the trigger transformer seems likely.

Peter van Hoof



{Original Message removed}

2009\07\17@084919 by Jinx

face picon face
> Is the trigger transformer close to the bulb?

http://home.clear.net.nz/pages/joecolquitt/flash_proto.jpg

I've super-imposed the tracks. It's a fairly simple layout to follow

> Any chance of arcing from the trigger wire

I don't think so

> Did you try to reverse the two main electrodes of the bulb?

No

> At this point a problem with the trigger transformer seems likely

The two things I haven't replaced are the transformer and the bulb. I
very nearly went into town today to buy 10 of each but I called ahead
and they couldn't help with the transformers, so I didn't go. I've just
the one of each therefore.

2009\07\17@110227 by Dave Tweed

face
flavicon
face
Jinx wrote:
> > Is the trigger transformer close to the bulb?
>
> home.clear.net.nz/pages/joecolquitt/flash_proto.jpg
>
> I've super-imposed the tracks. It's a fairly simple layout to follow

Are you absolutely sure you haven't confused the primary and secondary
windings on the transformer?

-- Dave Tweed

2009\07\17@113356 by Alan B. Pearce

face picon face
>> > Is the trigger transformer close to the bulb?
>>
>> home.clear.net.nz/pages/joecolquitt/flash_proto.jpg
>>
>> I've super-imposed the tracks. It's a fairly simple layout to follow
>
>Are you absolutely sure you haven't confused the primary and secondary
>windings on the transformer?

I haven't seen any reply to my take on it - I would still like to see a
resistor between gate and cathode to lower the sensitivity to stray hum and
EM stuff triggering the SCR.

2009\07\17@132714 by AGSCalabrese

picon face
I think you have more than one problem.


#1)     You are not seeing the proper voltages for triggering the  
SCR.   Your opto may be to blame.


#2)     You are not triggering the flash even when you dump current  
through the transformer.

If you are still interested, I can suggest methods for testing each of  
these.

Gus

2009\07\17@132917 by AGSCalabrese

picon face
>
> On Jul 17, 2009, at 9:34 AM, Alan B. Pearce wrote:
>
>>>> Is the trigger transformer close to the bulb?
>>>
>>> home.clear.net.nz/pages/joecolquitt/flash_proto.jpg
>>>
>>> I've super-imposed the tracks. It's a fairly simple layout to follow
>>
>> Are you absolutely sure you haven't confused the primary and  
>> secondary
>> windings on the transformer?
>
> I haven't seen any reply to my take on it - I would still like to  
> see a
> resistor between gate and cathode to lower the sensitivity to stray  
> hum and
> EM stuff triggering the SCR.

that will prevent false triggering.... it will not fix triggering not  
happening.
Gus

2009\07\17@140159 by Dwayne Reid

flavicon
face
At 08:49 PM 7/16/2009, Jinx wrote:
>Picked up my H11D1 optos for Figure 3 at
>
>http://www.rory.co.nz/projects/lighting/beat_strobe.html
>
>and been trying, unsuccessfully so far, to get a flash out of it
>
>The problem appears to be that the SCR gate rises to only 200mV
>when the opto is activated. Looking on the scope there don't seem
>to be any gate pulses higher than 200mV and no activity at all on the
>cathode. Voltage on the cold side of the 2k2 (opto collector) also
>drops to near-zero

It sure sounds like something is funky with the opto.  Try removing
the opto from the circuit, then measure the voltage across the
SCR.  It should be somewhere between 50 - 100 Vdc.  If so,
momentarily short the leads that used to connect to the opto output -
the strobe should flash.

If that does happen, look for problems with the opto.  Transistor
wired backwards?  Too much leakage?

If all else fails, do consider trying a MOC3011 instead of the opto
you are currently using.  Like I had mentioned earlier, I've built
MANY strobes using the MOC3011 as the trigger element (it replaces
the SCR).  All still working to this day.

dwayne

--
Dwayne Reid   <.....dwaynerKILLspamspam.....planet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(780) 489-3199 voice          (780) 487-6397 fax
http://www.trinity-electronics.com
Custom Electronics Design and Manufacturing

2009\07\17@180747 by David Duffy (AVD)

flavicon
face
Jinx wrote:
>> Is the trigger transformer close to the bulb?
>>    
>
> http://home.clear.net.nz/pages/joecolquitt/flash_proto.jpg
>
> I've super-imposed the tracks. It's a fairly simple layout to follow
>  
A couple of notes:
Those tubes are polarity sensitive I've found.
In my similar design, I used a 470p / 1KV cap in series with tube trigger
The transformer is incorrect - swap the two end connections
David...

--
___________________________________________
David Duffy        Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: +61 7 38235717      Fax: +61 7 38234717
Our Web Site: http://www.audiovisualdevices.com.au
___________________________________________

2009\07\17@184723 by Peter van Hoof

face picon face

I agree the two high voltage wires on the trigger transformer should be reversed.
Also usually you will find a red mark on the positive connection of the flash bulb.

Peter



{Original Message removed}

2009\07\17@190620 by Jinx

face picon face
> I don't understand why he has the 1n4007 in there either. this seems
> unnecessary.

My understanding is that the standard 1N4007 is not a particularly
fast diode (for example it's not suitable for semi protection in noisy
applications). If a diode is at all necessary there I'd have thought even
a small fast signal diode (1N4148) might cope, given the current
limiting of the 2k2 and the opto's transistor. I have some UF4004
and SMT fast suppression diodes on stepper boards

2009\07\17@193222 by Jinx

face picon face
> I would still like to see a resistor between gate and cathode to
> lower the sensitivity to stray hum and EM stuff triggering the SCR

What would you suggest ? I see quotes that some SCR in-built
resistors are as low as 20 ohms. Others suggest 100 ohms, eg

http://www.microsemi.com/micnotes/601.pdf

I was initially thinking a few k perhaps

--------------

Reading these two for a little education

http://www.allaboutcircuits.com/vol_3/chpt_7/5.html

electricalandelectronics.org/2009/04/13/thyristor-gate-characteristic
s/



2009\07\17@201240 by Jinx

face picon face
Peter,

> I agree the two high voltage wires on the trigger transformer
> should be reversed

Ah-ha, now we're cookin'

It's a curious case of self-deception. If you compare my layout with
the author's

http://home.clear.net.nz/pages/joecolquitt/flash_proto.jpg

vs Figure 6 at

http://www.rory.co.nz/projects/lighting/beat_strobe.html

they look superficially the same, which is why I made the fatal
mistake of not verifying where those tracks were going. HOWEVER,
his voltage rail tracks are the opposite polarity wrt the transformer
compared with my layout. So, I've re-orientated the transformer
(at this stage by simply putting it on the underside of the PCB) so
that the common is to the tube trigger, not the 0V rail where it was

Activating the opto now produces a nice fat flash

> Also usually you will find a red mark on the positive connection of
> the flash bulb

Can't see any external markings, but the terminals are different, so
I'll check up on tube construction. The tube does flash now, which
is the main thing

Disappointing in one way to make a silly mistake like that but if
it wasn't for (occassional) dimwits like me we wouldn't get to pull
circuits apart ;-))

2009\07\17@222545 by Jinx

face picon face
I wrote

> compared with my layout. So, I've re-orientated the transformer
> (at this stage by simply putting it on the underside of the PCB) so
> that the common is to the tube trigger, not the 0V rail where it was

Sorry, typo. The common was previously on the tube trigger, now
it's on 0V where it belongs. The hot end of the secondary is now on
the tube

2009\07\17@225812 by Peter van Hoof

face picon face

Credit goes to David Duffy he first spotted it I just mentioned it again
to make sure you read it.

Peter



{Original Message removed}

2009\07\17@232402 by Jinx

face picon face
> Credit goes to David Duffy he first spotted it I just mentioned it
> again to make sure you read it

Ah, yes. I see that now. Dave Tweed also mentioned it. Must be
a David thing



2009\07\18@141428 by Alan B. Pearce

face picon face
>> I haven't seen any reply to my take on it - I would still like to
>> see a resistor between gate and cathode to lower the sensitivity
>> to stray hum and EM stuff triggering the SCR.
>
>that will prevent false triggering.... it will not fix triggering
>not happening.

Well, I get the impression that something is permanently triggering the SCR,
like a nearby cellphone tower, or mains noise induced into the circuit, or
...

The gate will be a very high impedance point in that circuit, so anything
could trigger the SCR, and there will be only small voltages around it if
the noise is continuous.

2009\07\18@142708 by Alan B. Pearce

face picon face
>> I would still like to see a resistor between gate and cathode to
>> lower the sensitivity to stray hum and EM stuff triggering the SCR
>
>What would you suggest ? I see quotes that some SCR in-built
>resistors are as low as 20 ohms. Others suggest 100 ohms, eg

Hmm, in view of the method of driving, I would use somewhere between 1k &
10k, and do away with the 1n4007 entirely by shorting it out.

2009\07\18@204745 by Jinx

face picon face
> Hmm, in view of the method of driving, I would use somewhere
> between 1k & 10k, and do away with the 1n4007 entirely by
> shorting it out

Now that I've seen the circuit does actually work (if you put the
transformer in the right way. Moving along....) it does seem to
tempt fate having the gate just float. And I'm not sure about that
diode either

2009\07\18@212405 by David Duffy (AVD)

flavicon
face
Jinx wrote:
>> Hmm, in view of the method of driving, I would use somewhere
>> between 1k & 10k, and do away with the 1n4007 entirely by
>> shorting it out
>>    
>
> Now that I've seen the circuit does actually work (if you put the
> transformer in the right way. Moving along....) it does seem to
> tempt fate having the gate just float. And I'm not sure about that
> diode either
>  
Jinx,
My circuit has a 4K7 from gate to cathode and doesn't have the diode.
David...

--
___________________________________________
David Duffy        Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: +61 7 38235717      Fax: +61 7 38234717
Our Web Site: http://www.audiovisualdevices.com.au
___________________________________________

2009\07\19@194054 by Jinx

face picon face
David, do you think this tube is being driven conservatively ? If it's
the U6531 at this page

http://www.taytron.com/fttypeu.html

then it appears to have a 60Ws (joule) rating. The 10uF of the circuit
would be about 0.5J. Camera flash schematics I've seen, at eg

http://www.repairfaq.org/sam/strbfaq.htm

show anything from 250uF (12J) upwards, and 30Ws is what I've
read quoted for basic in-built camera flashes

Here's a quick comparison test, just to get some idea of the light output

http://home.clear.net.nz/pages/joecolquitt/flash_compare.html

Although the analyser traces indicate 10s of ms for the flash I'm
sure that's because the phototransistor is not being turned off
quickly enough

http://en.wikipedia.org/wiki/Xenon_flash_lamp

There certainly are some intricacies to protect the tube and capacitor

2009\07\19@213805 by David Duffy (AVD)

flavicon
face
The ones I built used a 22uF / 400V (or maybe 450V?) capacitor with a
similar tube. It was a few years ago so the exact details are a little
sketchy!
David...

Jinx wrote:
> David, do you think this tube is being driven conservatively ? If it's
> the U6531 at this page
>
> http://www.taytron.com/fttypeu.html
>
> then it appears to have a 60Ws (joule) rating. The 10uF of the circuit
> would be about 0.5J. Camera flash schematics I've seen, at eg
>
>  
--

___________________________________________
David Duffy        Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: +61 7 38235717      Fax: +61 7 38234717
Our Web Site: http://www.audiovisualdevices.com.au
___________________________________________

2009\07\19@215844 by Jinx

face picon face
> The ones I built used a 22uF / 400V (or maybe 450V?) capacitor
> with a similar tube

I've got 22uF on this now, which is a little over 1J. I'll track down
some disposable cameras - from a film processing lab probably -
and get their flash units / tubes for dissection and experimentation

2009\07\20@042058 by Alan B. Pearce

face picon face
>There certainly are some intricacies to protect the tube and capacitor

Normal photographic flashes extend the life of the tube by limiting the rise
time of the current by using specific lengths of wire between the capacitor
and the tube, thereby providing a suitable series inductance that limits
di/dt. Early model flashguns that used paper dielectric capacitors charged
to around 2kV could produce an extremely short and intense flash.

I have spoken to commercial photographers who used these old units and
considered it normal to be able to get an extremely sharp stop motion of a
ballet dancer doing a twirl, and in the photo the outer edge of the skirt is
absolutely crisp sharp in the photo.

later units that used electrolytic capacitors charged to around 500V do not
have the same short flash, as the internal impedance of the capacitor lowers
the di/dt. Also I suspect the lower operating voltage means the tube is
harder to fire.

The most modern units seem to run at around 200-300V across the storage
capacitor.

2009\07\20@065439 by Peter van Hoof

face picon face

Modern flash units do cut down the duration of the flash if not the entire
charge on the capacitor is needed, this preserves the remainder of the
stored energy in the capacitor and makes the flash duration shorter.

Always be careful building a strobe like unit and comparing it to camera
flash units as it is important to realize the bulb limits have to do with both
the peak energy in one flash and the total energy over time. (cooling
issues).

If you wish to raise the amount of energy at the same flash rate you also
may have to change the charge circuit.

Peter van Hoof



{Original Message removed}

2009\07\20@082933 by Jinx

face picon face
Peter and Alan

> Modern flash units do cut down the duration of the flash if not the entire
> charge on the capacitor is needed, this preserves the remainder of the
> stored energy in the capacitor and makes the flash duration shorter.

I went out today and picked up a bagful of used disposable cameras

http://home.clear.net.nz/pages/joecolquitt/disposable_camera.html

And plenty more where they came from

The capacitor is unmarked apart from T0440 (which one could guess
is the voltage rating). One of the caps still had 170V on it. I estimated
its capacitance by comparing the charge time compared with a marked
220uF

The capacitor is right across the tube terminals. The whole circuit is
pretty basic. It's a small tube, only 21mm long in total, with just 13mm
visible in the reflector. Looks like the shutter leaf switch sends a pulse
from the reservoir cap into the trigger transformer

The plan is to substitute this tube in the 22uF circuit and see what
sort of light is produced. Then substitute the 250uF for the 22uF,
then re-install the U-tube

It could be that the camera tube has an expected lifetime of something
a little over the 20 shots on the film. Or it may actually last much longer

2009\07\20@133216 by Dwayne Reid

flavicon
face
At 06:29 AM 7/20/2009, Jinx wrote:

>I went out today and picked up a bagful of used disposable cameras
>
>It could be that the camera tube has an expected lifetime of something
>a little over the 20 shots on the film. Or it may actually last much longer

I've used dozens and dozens of those disposable camera flash units -
haven't killed a tube yet.  I was using 10uF 350V electrolytic caps
with the tubes since what I needed was a repetitive strobe flash for
signalling, not just a single but intense flash.  So far as I know,
all are still working after many years of relatively continuous use.

As an aside, I've used the little DC-DC convertor circuit and that
large (330uF 350V) cap to make blasting boxes for some of the local
pyro guys.  The existing transistor was too wussy for continuous use
- I replaced it with a similar gain TO-220 device.  The battery I use
is a single "D" alkaline cell.  They typically get at full year's use
out of a single cell, using them several times per week.

The guys seem to like them.  They take longer to charge than the
commercial blasting boxes that they use but quite frankly, that's
never a problem.  One neat thing is that I tapped off the voltage
spike at the collector of the oscillator transistor and used that to
make a bog simple loop resistance indicator consisting of 2
LEDs.  One LED only lit: open circuit.  The other LED only lit: short
circuit.  Both LEDs lit equally: about 100R loop resistance.  Maximum
loop test current never exceeds 10mA under any circumstance which is
just fine for all pyro initiators that I am aware of.

100R loop resistance with >300Vdc is more than 3A firing current -
quite acceptable.  Maximum short circuit current is limited with a
large (20W) 10R resistor so as to not completely fry the switch
contacts when the output terminals are shorted.

dwayne

--
Dwayne Reid   <EraseMEdwaynerspam_OUTspamTakeThisOuTplanet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(780) 489-3199 voice          (780) 487-6397 fax
http://www.trinity-electronics.com
Custom Electronics Design and Manufacturing

2009\07\20@183541 by Jinx

face picon face
Thanks Dwayne

wbr

2009\07\21@035229 by Jinx

face picon face
> I've used dozens and dozens of those disposable camera flash units -
> haven't killed a tube yet.  I was using 10uF 350V electrolytic caps
> with the tubes since what I needed was a repetitive strobe flash for
> signalling, not just a single but intense flash.  So far as I know,
> all are still working after many years of relatively continuous use.

Some rough comparsions of tube outputs with two different caps
and stops

http://home.clear.net.nz/pages/joecolquitt/tube_compare.jpg

2009\07\23@182951 by Jinx

face picon face
> I have spoken to commercial photographers who used these old units
> and considered it normal to be able to get an extremely sharp stop
> motion of a ballet dancer doing a twirl, and in the photo the outer edge
> of the skirt is absolutely crisp sharp in the photo

I've had a first go at measuring the flash lengths. The length of the bar
is 160mm (describes a 1m circumference circle), rotated at 2400rpm
-> 40rps -> 40,000mm/s

In the first there's little blurring. Flash time must be quite short as 1mm
blur would be 25us. I've enhanced the end portion. What I should have
done is have something more definable (a drawing perhaps) at the end
of the bar to test for blurring. The second shot is definitely blurred, even
without enhancement, possibly 50-100us flash. Although the more
powerful flash may be reflecting off the walls enough to blur the shot

The 22uF flash would probably be adequate with a wider aperture and
some shot processing. 47uF might be a fair compromise

http://home.clear.net.nz/pages/joecolquitt/speed_test1.jpg

2009\07\24@080445 by olin piclist

face picon face
Jinx wrote:
> The second shot is definitely
> blurred, even without enhancement, possibly 50-100us flash. Although
> the more
> powerful flash may be reflecting off the walls enough to blur the shot

I'm confused about what exactly you're trying to accomplish, but bouncing
off walls does not meaningfully extend the illumination time.  At 11.8
inches/nanosecond, any bounces will be too short lived to measure with this
setup, and tiny compared to the flash length.  If the round trip bounce
distance is 30 feet (about 10 meters for the imperially challanged), then it
adds only about 30nS to the total illumination time.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2009\07\24@214139 by Jinx

face picon face
> > powerful flash may be reflecting off the walls enough to blur the shot
>
> I'm confused about what exactly you're trying to accomplish

Just a simple set-up to get some idea of how bright different flash tubes
are and how the value of the storage capacitor affects flash length

> bouncing off walls does not meaningfully extend the illumination time

Yes, that's true. This was in quite a small room, barely 3m square. What
I intended to suggest was not time effects, but whether the walls act as
reflectors and add enough to the illumination to cause more visible blurring


'[EE] High voltage opto-coupler'
2009\08\13@061520 by Jinx
face picon face
> I've used dozens and dozens of those disposable camera flash
> units - haven't killed a tube yet.  I was using 10uF 350V caps

Things progressing nicely with the project and initial tests look
promising. Hopefully should be able to share soon

Couple of questions -

----------

Are there any particular parameters to look for when selecting
the storage capacitor ? For example series resistance. The flash
tube draws a lot of current for a short time and I worry that this
is like the screwdriver method of discharging a capacitor, which
I believe is not good for it

-----------

I'd like to have the capacitor selectable for different flash rates
and different brightnesses. For instance select between 22uF, 100uF
and 220uF. Refering to Figure 3

http://www.rory.co.nz/projects/lighting/beat_strobe.php

I have some toggle switches rated at  250V 3A / 125V 3A. As
they're DPDT, the contacts can be paralleled for twice the current

Do you think a switch between capacitor -ve and the 0V rail will
be OK ? The author of the page above calculates that the current
through the tube is of the order of > 500A/cm^2, and I reckon the
duration is around 25us for 22uF and 100us for 250uF

What I'm not sure of though is the current in the -ve rail, ie whether
amperage seen in the tube would also be apparent through the switch.
The static voltage across the switch will near 0 when idle, going by
the datasheet resistance of 10 milliohms

If it might take out the switch then perhaps a power FET between
the -ve of each capacitor and the 0V line. The FET gates can be
controlled with a low-power rotary switch to the gates. The cost
would probably be similar compared with a toggle switch as above

TIA

2009\08\13@193036 by David Duffy (AVD)

flavicon
face
On the ones we built (some time ago), I don't think we had any capacitor
failures with the standard 450V, 105 degree types we used.
David...

Jinx wrote:
> Are there any particular parameters to look for when selecting
> the storage capacitor ? For example series resistance. The flash
> tube draws a lot of current for a short time and I worry that this
> is like the screwdriver method of discharging a capacitor, which
> I believe is not good for it
>  
--

___________________________________________
David Duffy        Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: +61 7 38235717      Fax: +61 7 38234717
Our Web Site: http://www.audiovisualdevices.com.au
___________________________________________

2009\08\13@201200 by Jinx

face picon face
> On the ones we built (some time ago), I don't think we had
> any capacitor failures with the standard 450V, 105 degree
> types we used

Thanks. I'm looking through datasheets but not finding anything
directly obvious. There's some mention of special capacitors for
high-energy studio equipment but I think standard photoflashes
are well short of that sort of discharge

2009\08\13@202145 by David Duffy (AVD)

flavicon
face
Jinx wrote:
>> On the ones we built (some time ago), I don't think we had
>> any capacitor failures with the standard 450V, 105 degree
>> types we used
>>    
>
> Thanks. I'm looking through datasheets but not finding anything
> directly obvious. There's some mention of special capacitors for
> high-energy studio equipment but I think standard photoflashes
> are well short of that sort of discharge
>  
I do remember wondering if it was going to be a problem, but I think
there were only tube, trigger transformer (and its coupling cap) failures.
David...

--
___________________________________________
David Duffy        Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: +61 7 38235717      Fax: +61 7 38234717
Our Web Site: http://www.audiovisualdevices.com.au
___________________________________________

2009\08\13@203623 by Jinx

face picon face
> I do remember wondering if it was going to be a problem, but I
> think there were only tube, trigger transformer (and its coupling
> cap) failures.

As those components are being scavenged from disposable cameras
there's an endless free supply of those. The prudent thing to do would
be use a smaller cap than the original to put less-than-designed stress
on them

2009\08\26@074511 by Jinx

face picon face
I mostly finished the initial boards and s/w a couple of days ago
and tried it out briefly

http://home.clear.net.nz/pages/joecolquitt/multi-flash.html

There's no text to that page yet. Parts used from the cameras
are the tubes, trigger transformers and 22n cap. The flash section
circuit is based on the link I've posted before. I found the camera
trigger transformer primaries need much more than the 65V of that
circuit, so I swapped the 100k/390k divider. Luckily I did order
and have the 300V optos

Results would be better if a non-reflective black backdrop was
used and/or the background was much further away. The glove
example clearly shows how the flash reflected from a light back-
ground causes apparent transparency

I think the flash is bright enough at close range with 22uF. The
original cameras have anything from 80uF to 200uF. I'm thinking
of making a second set of single flashes with selectable caps.
Probably will add a switchable camera 80uF in parallel with the
22uF. Fortunately the 325VDC is just under their 330V rating,
but I might be happier if the DC was perhaps closer to 300

Capturing the water-filled balloon bursting was beginner's luck, I
tried a blowgun just the once. A pin with cotton wool wadding
was blown through the IR beam, after checking the timing with
a dummy shot

More... (looser matching)
- Last day of these posts
- In 2009 , 2010 only
- Today
- New search...