Page 2 of 3
Posted: Sat Apr 24, 2010 1:54 pm
by jimmy101
mark.f wrote:
Now that you mention it, it might be feasible to simple replace the 120uF photocap with the smaller trigger cap, instead of ordering another capacitor (Radioshack doesn't have the best selection in metal film).
The high power cap really doesn't need to have all that great an ESR. I've looked and caps like the trigger cap are pretty rare. Even places like Digikey and Mouser don't have that great a selection of very low ESR caps.
But, the power cap really doesn't need all that great an ESR. Photocaps aren't anywhere as good, in terms of ESR, as are the triggers caps. The photocap doesn't have to be since an extremely high discharge rate really isn't needed. The thing that is special about photocaps is that they can be repeated charged to high energy densities then basically short circuited. Most high power caps really don't like that. The big-ass filter caps in a power supply for example won't last long if they are constantly being charged and short circuit discharged.
So, almost any cap can be used as a replacement for the 120uF cap. If the capacitance is smaller then it'll last quite a while. A 1.2uF cap will charge 100X faster the the 120uF but it'll only contain 1% the energy.
People have hacked these photoboards to auto-repeat. They replace the photocap with a uF or so cap and replace the fire switch with a couple small neon bulbs in series. Small neon bulbs breakdown (light up, a spark jumps the electrodes) at about 70V. Put a couple in series and you get a switch that auto triggers when the trigger cap voltage gets above 140V (two Ne's), 210V (three Ne's) ...
Posted: Sat Apr 24, 2010 3:15 pm
by mark.f
That's pretty interesting. I was looking at an SCR with a high-value resistor wired in series with the gate before you mentioned the neon bulbs. That's much simpler.
I do believe my control box is taking shape. It's also gotten a lot simpler in circuitry since I started this thread, which I thank everyone for. Next paycheck I can start buying basic cannon materials.
Posted: Sat Apr 24, 2010 4:40 pm
by Hotwired
It'll be difficult to put enough trigger caps in parallel to get sparks capable of igniting anything reliably. Can't remember how much they store but not a lot and its in pico farads.
I've found quite a few suitable caps on digikey and mouser but dam they have a nasty minimum order quantity.
Posted: Sat Apr 24, 2010 9:20 pm
by Insomniac
Just a note on neon bulbs... I've used them in the past to make an auto-repeating flash, and never had any sucess. The voltage builds, the bulb flashes orange, then starts to slowly flicker at a gradually increasing rate. Seemed to have too much resistance to allow the trigger cap to actually trigger the flash.
Also tried a method using a globe and an SCR, which did't work either. However I've seen this method used successfully so I'll put my failures down to my minimal electronics knowledge.
Posted: Sat Apr 24, 2010 9:33 pm
by mark.f
Well, I made my first prototype tonight.
I tried the neon bulbs, but I guess I couldn't get a hold of the right ones, because with FOUR of these in series, they would be powered with no problems just from the rectified DC from the transformer.
So, in a ditch effort to make it work, I rigged up a spark gap for the capacitor, with the electrodes simply being solder-coated leads of wire.
When I FINALLY got the spark gapped correctly, it sparked at a rate of roughly one spark per second (this was with a 1 uF capacitor).
Here's a picture below. This was the first serious circuit I've put together, by the way...
Posted: Sat Apr 24, 2010 10:33 pm
by Hotwired
I did say about those manual spark gaps
Seems to be an extra transistor/thyristor/3pinnedwossit along with two extra resistors on the board?
Shouldn't need more than the transformer + resistor + transistor + diode with optional extra resistor + led for charging status.
Click for ye olde circuit sketchings:
can't be bothered making it more legible
Actually you might be using a different transformer which would make more sense. The small 5 pin green/blue transformers primarily from Kodak boards are the ones I use. Some Japanese brands and generic types for sale in say wedding bulk packs use up to 7 pin transformers.
Posted: Sat Apr 24, 2010 10:48 pm
by mark.f
There's a voltage regulator and three extra resistors (to adjust the voltage regulator... didn't have a 20 ohm resistor so I combined a couple...). This fits more closely the final circuit (where two 9 volts run the ignition and fan).
I might actually just try using an SCR to switch the current to the coil. With a big enough resistor it should supply the gate current at the full charge voltage...
If not, I could just make it a "one shot" igniter.
EDIT: My transformer has the same configuration as yours (5 leads, each to the same type of internal coil), but the pinout is just different. I check the pinout and mark it down before removing the transformer from the PCB.
EDIT II: So do you think this configuration would work? The voltage divider should supply 1.5v (the gate voltage to the SCR) when the capacitor reaches 250 volts...
Posted: Sun Apr 25, 2010 1:03 pm
by jimmy101
You might want to take a look at the instructable at
http://www.instructables.com/id/Hack-a- ... be-light-/
or
http://www.josepino.com/light/strobe_from_flash
These use the NE-2 bulb (the board's "ready" light) to trigger the SCR. That might work a bit better than a voltage divider since the voltage divider places a pretty significant load on the trigger cap charging circuit. Often the trigger cap (20ish nF) is charged via a one to five megaohm resistor. That means there is very little current and almost any load will sink more ooomph than the megaohm size resistor can supply.
The megaohm resistor is what keep the main photocap from discharging through the trigger coil.
Posted: Sun Apr 25, 2010 1:30 pm
by Hotwired
Argh, had to register to see what the pictures were, only lets you see thumbnails if you're not.
Looks possible.
I had a tinker:
This one skips the trigger stage because without a flash bulb there is no need for it (if your SCR can't deal with the current then there's an argument for keeping the triggered flash tube).
As soon as the voltage is sufficient to light the bulb the SCR triggers and dumps the capacitor through the primary.
Posted: Sun Apr 25, 2010 9:03 pm
by mark.f
That would work too. I don't know a terribly good deal about SCR's, but I think they'd only open when a certain voltage/current was reached on the gate, which is why I inquired about the voltage divider.
I can't find a suitable neon lamp anywhere, but since I have to order an SCR I'll see what Mouser has...
Posted: Sun Apr 25, 2010 9:18 pm
by rp181
SCR trigger currents are VERY low. I could trigger a large 5000A one with ~2v.
Make sure you drive the SCR gate hard (add a capacitor in parallel with your trigger PSU). If the gate signal is too weak, the gate doesn't open fully/opens slowly, and causes more heating.
Posted: Sun Apr 25, 2010 10:56 pm
by Hotwired
It should work, the main difference between this and the demonstrated working circuit on Instructables is that the SCR is completing the circuit for a larger capacitor.
The limitation seems to be that the neon bulb will conduct at a rather lower voltage than even the basic ~300V of the original circuit. Perhaps 100V - need to check something about that actually.
I've got a few thyristors about, blew several trying to switch the large photo caps in the hope that I could sneak past the 5A rating on them
Think I'll dig through the pile of camera circuits I have, see if there's any with neon bulbs on*
May try with a smaller foil cap once I've rebuilt my circuit.
*A very brief check later and it looks like almost half the camera boards I have use a small glass neon bulb instead of a small LED as a charging indicator.
Posted: Mon Apr 26, 2010 12:10 pm
by Technician1002
rp181 wrote:SCR trigger currents are VERY low. I could trigger a large 5000A one with ~2v.
Don't confuse voltage with current. The trigger voltage is under 2 volts, but some current is required. Unless you are using a sensitive gate SCR, a neon lamp might not trigger it. Check the spec sheets of any parts you intend to use. A solid pulse of several mA is enough to trigger most SCR's. A few microamps is too little and will need some sort of trigger circuit to provide a pulse. A diac and capacitor work well. Charge the cap until the diac breaks over and triggers the SCR is a common arrangement used in light dimmers. Diac voltages in about 20 volts are common.
Posted: Mon Apr 26, 2010 5:01 pm
by mark.f
Yes, the SCR I intend to use has a trigger current of 25 mA. I have to order a few parts from Mouser so I'll see what I can cook up...
Posted: Thu Apr 29, 2010 5:43 am
by mark.f
Well, I'm tired of trying to make this circuit repeating. Maybe some other day, but for now, I'm just making it a "single shot" ignition system.
Got my SCR's in from Mouser, along with some T-2 neon bulbs (I assumed these were the TE-2 ones jimmy was talking about). With five of them in series they glowed bright orange when connected to my prototype circuit (should take around 450 volts to break down... so...).
For now, this is my circuit design. I'm going to try to lay it out on perfboard. It needs to be small enough to fit into 2" PVC along with some other parts. Got everything taped down to the perfboard now, just adjusting the positions to get it as compact and easily jumpered as possible. Switches and LED's are run off the board for now with simply leads.