Strobette Stroboscope

[GoNe_aWoL]

Will this work? Modeled off a "Strobette Model 964 Stroboscope" my UJT builds heat whenever power is on, the 2646 xrefrence NTE6491 at $6.50 is too expensive to keep burnin' up. I've used a neon between the emitter and B1 and veirfied that it will flash, but it's inconsistant. This light was built to vary between 200 and 6500 flashes/min. to check/confirm speeds of moving parts. I'm expecting that the UJT is somehow incorectly placed in this schematic.
I'm hoping someone can help me out. All I'm trying to accomplish is a light to do what this was intended to do.

 

[Nigel Goodwin]

I would suggest D4 and D5 should be zener diodes, you're probably applying far too much voltage to the unijunction and blowing it!.

 

[GoNe_aWoL]

Duh! Thanks Nigel, you've inspired my new signature. :?
I makes sense to me (atleast for now) that D4 should prevent C6 from overcharging seen by the emitter, and D5/R8 can regulate the voltage seen by B2. What values for D4/5? R9 is a 10W and R4 is 1/4W, I have a datasheet on the UJT, but I'm not sure I understand fully what it's telling me. I want to try a pair of 25V, I think that would prevent anything from frying, But that seems high for a little tin can like the 6401. C6 has no voltage limit for me to start from, It's a wonder I havn't popped a couple of them yet.
Any replies are very much appreciated!

 

[Nigel Goodwin]

If you've 'modelled it from a Strobette Model 964 Stroboscope' does this mean you have access to one?. If so, simply measure the voltage across the zeners.

If not, I would try something like 15V or 12V.

 

[Roff]

Edit:
This post contained erroneous info. See my post below.

 

[Optikon]

I don't know if your circuit is wrong, but I believe your schematic is. It doesn't work the way it is drawn (I sim'ed it to check my suspicions). I'm betting that C1, C3, D2, and D3 are meant to be a voltage doubler. See the corrected schematic below.
All these diodes need to have at least 400V breakdown voltage.

What did you use to simulate this? If I may ask..Is that a tube in there? Who models those these days?

I wouldnt be surprised if you can do this in SwitcherCad, its a darn fine simulator!

 

[GoNe_aWoL]

Nigel,
Thanks again for your advice! I actually drew this schem (7+ years ago) from one I have, I don't remember what D4 and D5 were, but I had replaced them, I don't remember why.

Ron_H,
You've nailed that one... I remember at the time that ofcorse it was a vx but aparently didn't review the schem. C1,C3 are infact 500v, c2 being 450.

Opitkon,
lol, I couldn't find a stobe bulb in circut maker, so I improvised! Thanks for telling me about SwitcherCad, I've downloaded and have started playin with that one...

I'm not as swift as I was, and I didn't know what I was doing then either~
Based on that, I want to thank everyone for putting up with such a BIG-little prodject.

 

[Roff]

I screwed up. The original power supply section is OK, except change C2 to 220uF. The high voltage output is a voltage tripler, about 450-500v. It won't work with C2=20uF. I'm going to delete the schematic I posted above.
Optikon, I only simulated the power supply section. I doubt a model of the strobe tube is available anywhere. I did simulate it on SwitcherCAD III.

 

[GoNe_aWoL]

Ron_H,
Your schem IS what the 964 is. Infact, I saved it over my own! C2 works like it should, holding @150V supplying juice to the emitter, I'm testing C1 at 330, C3 at <500. (with my $10 kmart meter).
I'm planning a trip to RS to find the zeners today for D4 and D5, what are your thoughts on those?

 

[GoNe_aWoL]

I've got to take that back. I was tired and was too hopefull for a breakthrough. I've learned my lesson... No more contradicting "Electronics gods" Now I'm seeing what Ron was saying. I downloaded SwitcherCAD, but ran into trouble finding a UJT, then found the Student CM that I'd downloaded long ago.
I've examined the scope and reconfirmed the schem I posted in the first post. C2 has thrown me for a loop, It's the original, it says:
Mallory
20 MFD
450 VDC

Was capacitance ever measured in Mili farads? The can itself is bigger than the 22uF 500V C1 and C3.
I've been thinking back, trying to remember little details I've forgotten. I've also gotten re-aquainted with circuit maker and have been running sims on this for awhile now changing D4 and D5 values, as well as connections of R10 and R11.
I've changed c2 to 200uF and I like the speed and smoothness the voltage reaches in the circuit. Ofcourse the UJT fires, but I can't seem to find a combination that'll give me the 200 - 6500 flashes a min. that this scope is supposed to produce.
Another key component I'm questioning is the supply voltage... What would simulate a north american 60hz sine-wave household outlet?
Anyway, thanks for all the help and I'm still workin' on it.

 

[Nigel Goodwin]

I've examined the scope and reconfirmed the schem I posted in the first post. C2 has thrown me for a loop, It's the original, it says:
Mallory
20 MFD
450 VDC

Was capacitance ever measured in Mili farads? The can itself is bigger than the 22uF 500V C1 and C3.

It 'could' be, milli is a perfectly correct prefix, but as far as I'm aware it never has been used - 22Mfd is 22 Micro Farads - I was always of the opinion that this labelling convention came from the USA, but I'm open to any other ideas?.

I've changed c2 to 200uF and I like the speed and smoothness the voltage reaches in the circuit. Ofcourse the UJT fires, but I can't seem to find a combination that'll give me the 200 - 6500 flashes a min. that this scope is supposed to produce.

That sounds rather fast!, are you sure!. Bear in mind the storage capacitor has to charge between flashes - this takes a fair amount of time.

Another key component I'm questioning is the supply voltage... What would simulate a north american 60hz sine-wave household outlet?
Anyway, thanks for all the help and I'm still workin' on it.

What mains do you have?, there's no reason to bother with 60Hz, 50Hz will work perfectly well - and if you have 230V mains you don't even need the doubler!.

 

[Roff]

Here's something to try. I changed R6, R10 and C6, and got rid of R5 and R11. I changed the symbols for the Zeners, and changed the value of C2.
Let us know if it comes close to working. :roll:
You might want to play with the value of C6 to get the range you want, as a UJT is spec'ed rather sloppily, and my spice model is questionable.
I think R4 could be 10k, saving a little power.
The R2*C5 time constant will recover to about 90% of full voltage in 10ms, which is about the minimum flash period you want, so it might work at 6500 flashes/min.