With a 3 meg resistor. That dissipates 8.3 watts. Your requirement was to operate at 48 volts and dissipate 1 watt in the bleeder. That gives 25 megohms. Your circuit will not work below 110 Volts with a 25 megohm series resistor. It hangs.
that
LED
avoid
No. It hangs. See the ASC file I posted. It produces one pulse and then hangs.
The risk with a hang is the circuit can fail with changes in component tolerances, transistor beta, temperature, leakage, etc. You stated you would rather not have a circuit that hangs. This is very solid engineering wisdom.
I showed you that yours does.
voltage
LED
You are changing the rules. Of course, you can simply make them anything you wish and declare victory.
Unfortunately, that destroys your credibility. Try convincing everyone who contributed their time and effort to the thread to do so again.
You stated 5KV operating voltage, 1 watt in the bleed resistor, no hang states, and works down to 48V. Also, although not stated explicitly, it must be visible and have a reasonable rep rate. That would be common engineering.
What kind of bs is that. You require a cap to store the charge so you can dump it into the led.
How are you going to get sufficient brightness to warn a tech as the HV decays down to 48V.
You criticised the contributions of others on circuit hangs and insufficient brightness.
Please apply the same criteria to your own circuit.
I doesn't hang, it just stops. Any LED driver eventually stops at some low voltage. Jim's circuit could hang, namely get into a nonfunctional, non-self-recovering state.
I'm the design engineer here, so I can change the rules.
-- John Larkin Highland Technology, Inc jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
It locks up. You have little or no control over the hang state. It fails long before meeting the minimum requirement of 48V.
Try it and see how many are willing to help you next time.
And my circuit doesn't hang... all that is hung is Larkin's ability to understand the circuit >:-}
But that's nothing unusual. You might as well give up, Steve, Larkin will dance and dodge all around and BS you to death. ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
It doesn't hang so much as it never worked at all. The schmitt was designed to fire around 12 volts, and with your input current, it never gets above 5. It works with my 3.3M resistor, as it was designed to.
I said in the other thread that I may change the HV supply to a lower voltage. It looks like I can. But this is a discussion group, and only jerks come here for "victory." They should join the Marines if they need "victory."
If people find this problem interesting, they can play with it. If not, they can go watch television or something.
-- John Larkin Highland Technology, Inc jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
I explained your hang state, and you never commented on it except to deny it and fling your usual insults.
-- John Larkin Highland Technology, Inc jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
You changed the input resistor ftom 3M to 25M, so it doesn't work the way I designed it to. You broke it and now you want to blame me.
-- John Larkin Highland Technology, Inc jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
You didn't "explain" it, you hand-waved it. ...Jim Thompson
This subthread has been terminated due to lack of informative content.
ALL responses will be automatically deleted by Agent's filtering system.
GND | 5MEG 100n | ___ || +--+----------. +-----+|___|+---+-----+--------||+--+------+ NE-2 |
0..5K source | | || + '----------' z + | or a LED circuit 33V Z A V C| | - C| 1mh +---+/|2 tran SCR C| | + + .-. | | | | | | 200k | | | + '-' | GND | + +-----+ | | +ND (created by AACircuit v1.28.6 beta 04/19/05You will need a load on there because the L kick back places - volts back to the zener and hits the gate circuit hard if you don't. But the idea is to make sure it turns off, which it deos. I used the PNP+NPN combination to form a SCR instead of a real SCR. 36uA is the max load at pulse time from the 5K source, this is all the way down to 33Volts. Output on one side of the Cap is higher than the other, take your pick.
-140 or around -100 volts. NE-2 should work on either side, LED on the output side. Jamie
Due to chicken-heartedness, more like it.
I stated the current, the voltage drop across the PNPN pair, and why it was latched. I speculate that, once latched, increasing the loop current won't unlatch it, because the overall feedback is negative.
Hiding behind your porous killfile again?
I sure wish you would figure out how to get a killfile to work.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation
What unlatches the SCR?
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation
Sure, but it doesn't depend on another capacitor to make things work, like Jim's circuit does. If the input current can charge the storage cap to 12 volts, it flashes. If the current is less than a few mA, it has to release. No hang states, nothing latched up.
Dumping 10 volts from 1 uF into a white Cree makes a big fat flash.
Well, they did hang. And I subsequently quantified how much energy a good white LED needs to make a good flash.
Please make an OT contribution. Come up with a better circuit or something.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation
I changed the input resistor to meet your specification. With your original requirement of 5KV, a 25 megohm resistor will dissipate 1 watt, per your requirement.
As I stated, a 3 megohm resistor will dissipate 8.3 watts at 5KV.
This is outside your stated requirement. I did not break your circuit. You did.
As I showed, your circuit has a hang state. It produced one pulse, then locked up. You have no idea what causes it, what you can do to prevent it, or how to ensure the problem is solved.
I suspect a beta issue, and certainly component tolerance and temperature are strongly involved. The problem is unfixable.
I have been baffled for a long time on how you can know all the buzzwords to electronics, and yet fail so dramatically in actual comprehension. I have seen you come within microinches of a true breakthrough, then ricochet off in some other direction. I can name three separate issues where you failed to look past your nose.
Another example is not recognizing Joerge's suggestion of the NCP302.
Here is a device that has a built in voltage reference, a decent comparator, and flexible output configuration. You could fire a single shot when the input reached the threshold, and be certain of excellent repeatability and absence of hang states. It would solve your problem.
But now things are becoming clearer. You studied physics, which is a huge field. You know all the buzzwords, but you just skim across the surface of everything. You do not go deep into a subject and truly absorb it.
In this example, electronics is just another broad field that you barely touch on. You do not feel it, or know instinctively which is the right or wrong direction to solve a problem.
A minor example is posting your original circuit, using a sine wave as the driving voltage. How on earth you could make any measurements from this is beyond me.
Of course, you fixed it in the next version, but why did you do it in the first place.
Anyway, the problem is largely unsolvable. You are unlikely to get enough light from the led at 1 Hz to tell when the discharge is clearly finished.
Using the controller to monitor the voltage decay, and a simple gravity switch for security are far better engineering solutions.
But of course, you will never use someone else's suggestion. That is not part of your religion.
I have John. I recognized flashing a led from the 5KV would not give enough brightness at the end of the decay period.
I suggested instead to monitor the decay voltage with the control circuit, and flag an error if it did not meet spec.
I also suggested to make the access through the bottom of the unit, and use a gravity switch to kill the high voltage when the unit is tilted on its side or upside down.
You never bothered to reply.
Look at fig 5 of the data sheet. Where is that current going to come from?
That's the problem with CMOS schmitt triggers; they draw a spike of Vcc current near the switch point.
I had a few physics courses, high school and two semisters in college, but my degree is BSEE.
You know all the buzzwords, but you just skim across the
Check out my web site.
Apparently so.
So, my people have to work on it upside down? Like on car repair crawlers.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation
L unlatches the SCR.
P.S. You'll need a R driving the NE-2
Jamie
btw, I have another diode in series with the ZENER to block the
- volts to the gate circuit. Also, I made a SCR via a PNP and NPN and on the NPN base I have a 10k R driving it at the divider of the zener and common R. I am sure you know those practices to protect the transistors.. bases really don't like to be zenered and over driven that much!
Jamie
There is at least one circuit that satisfys all those needs using common parts none costing more than cents - the problem being it uses a lot (40) of those parts. Reasonable parts count is also wanted.
piglet
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.