PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2017-06-27, 13:14:49
News: If you have a suggestion or need for a new board title, please PM the Admins.

Pages: [1] 2 3 4 5 6 7 8 9
Author Topic: Circuit sj1. Terse and Technical only.  (Read 93504 times)
Group: Moderator
Hero Member
*****

Posts: 2106
For various reasons, asking that all posts in this thread be Terse and Technical.  Q's here OK if terse, or by PM.  Eschew dialog.    No term "C OP" or "O U" pls. Pls replicate.

Vin 2.5 V AA's
Rb 51K
MPS2222
C-B 151 pF
D red LED
L-B, L-O bifilar 9turns, ferrite toroid 1"OD,  ~90uH each
Ro 0
Rr 0
Cout 10,000uF @16V, disconnected normally
I(t) by V over 1ohm CSR's

Then

DVM 5-7: ~zero  (DSO shows Vpp 12.7V, f ~ 1.4MHz)
DVM 5(red)-6(black) + 5.1V  (DSO: Vpp ~5.1V), Co charges to ~4.8V in ~5 secs when added
DVM 6(black)-7(red) +0.6V  (DSO: Vpp 3.4V)  Dim red.

Reps should show ~same as first step.


« Last Edit: 2011-05-29, 14:44:51 by PhysicsProf »
   
Group: Moderator
Hero Member
*****

Posts: 2106
To
Ro 9.8K
Rr 3.1ohm
Rb 2K
No  Co
I(t) measured over 1ohm CSR's
Then see Tek3032 attached, Pin left red waveform, Pout right.
« Last Edit: 2011-05-20, 04:45:52 by PhysicsProf »
   
Group: Moderator
Hero Member
*****

Posts: 2106
For Pin, DPO gnds @2 (see schematic); Photo
Probes always x10
« Last Edit: 2011-05-20, 14:35:37 by PhysicsProf »
   
Group: Moderator
Hero Member
*****

Posts: 2106
Ferrite toroid 1"OD, 0.5"ID, 7/16" high, electronic goldmine G6683 or whatever you've got - try it.
Insulated Cu, 22 gauge (20-23). Now fun part: Bifilar winding CW, see attachment, 9 (+/-) turns.
 Combined pair to point 7, then one wire to C-B, other to point 4 (schm).  O0
« Last Edit: 2011-05-20, 22:09:11 by PhysicsProf »
   
Group: Moderator
Hero Member
*****

Posts: 2106
Contrast circuit sj1 with "standard" Joule Thief circuit, attached, JT has L's BEFORE xistor, and has much higher P consumption (Pin).
sj1 has LB-CB tank circuit, opens&closes xistor gate rapidly, sets freq~.  Also RB and Rr limit current flow, thus limit Pin in sj1 circuit.
Goal is small Pin (and large Pout in closed "OUT" portion, with Pout dumped on [R-D or C-O] =load).  OUT circuit loop allows larger current flow there. Diode (or LED) direction in post 1 is correct.
Attached scope shots show Pin (green P waveform oscillates around 0, mean~small) whereas Pout integral clearly non-zero -- that's the pattern seen with sj1.
Two examples of Pin for sj1 type circuits attached; one shows Pout waveform comparison (Vin @ 2.92V); other shows mean Pin values around 1.3mW (Vin@2.06V).
OTOH, JT circuit typically shows saw-tooth waveform -- much different.
   
Group: Moderator
Hero Member
*****

Posts: 2106
Tek3032 displays for sj1 type circuit @2.06Vin.  Cf Pin and Pout waveforms (red) and values.  Different wound-toroid and circuit board from reply#1 above; replication effciency results tend to corroborate.
« Last Edit: 2011-05-20, 14:46:06 by PhysicsProf »
   
Group: Moderator
Hero Member
*****

Posts: 2106
  I welcome technical comments.  O0

Schematic repost, easier to see JPEG I think.  Can someone re-do this schematic with SPICE or equiv? thx, much obliged.
Editing:  argh, on re-checking everything, I found one probe on 1X rather than 10X -- moved inadvertently, somehow.  
So -- here are corrected results = with both probes on 10X

DVM shows potential low is at pt. 6 in circuit.  So I attached grnds of ATTEN DSO x10 probes to pt 6.
Vin 2.5V, Rr=0, no Cap Co.  (Different build than in post #1)
Channel 1 probe to pt 7; across LED, yellow waveform.
Channel 2 probe to pt 5; across Ro = 9.7K, blue waveform.

Results interesting, displayed on attached (sorry a bit fuzzy); note that ch1 crosses zero (I added black dashed line at V=0) goes flat, and note large voltages:
Ch1 (LED) rms 2.56 V, pp 6.16, mean 1.84
Ch2 (Ro)  Rms 3.00 V, mean 1.60
      
« Last Edit: 2011-05-20, 23:45:36 by PhysicsProf »
   
Group: Elite
Hero Member
******

Posts: 3052
It's turtles all the way down
Professor

I have been following this thread with interest. You stated your goal in one of the posts.

Quote
Goal is small Pin (and large Pout in closed "OUT" portion, with Pout dumped on [R-D or C-O] =load).

With this in mind, as an old hand at switchmode design, can I ask a question and make a few recommendations ?

1) In light of your goal, what function does Rr perform? You state that it is used as a current limit. Considering this, it will certainly waste power.

2) Consider operating at higher Vin, where semiconductor Vdrop will be less significant. You will need to readjust bias.

3) Use a good quality Schottky diode rather than LED. Determine the sweet spot value for the output load resistor. The idea is to keep current through the diode low so as not to waste power there, unless you want to calculate that power and use it as part of the output consideration, even though it is unusable except as a heater.

4) With the output capacitor connected, there will be high peak currents in the diode. Best to not use it unless you need a filtered output. A small choke will limit peak current if it Cout must be used. Consider synchronous rectification for lowest loss in the rectifier device.

5) Readjust turns ratio so that power is not wasted in unnecessary base drive. There is nothing magic about a 1:1 ratio. Personally I like to use the heaviest gauge wire  I can for the main winding, while leaving enough room on the core for just enough turns of a thinner gauge to sustain oscillation. Since any decent transistor will have a current gain of at least 100, keep the ohmic losses in the main winding as low as possible.

6) If you can operate above 5 volts, consider using a FET instead of BJT, as the drive requirements will be lower.

7) If size is not a problem, use a larger core with same number of turns. The operating frequency will be lower, but this will mean less switching loss overall. Generally if you can keep the operating frequency low, core loss and transition loss will also be reduced. Judging from the waveforms in "try3" scope shot, I would hazard a guess that you have lots of switching loss from the rise and fall slopes.

Finally, I see nothing magic about the common collector configuration vs. the common emitter configuration of a standard blocking oscillator. The only slight advantage may be that base drive also appears in the inductor ramp up current, but this is two orders of magnitude less important than other lossy circuit problems.

Those are my first cut suggestions. While some of these suggestions are more applicable to power supply design in the 10's of Watts range, if you are really trying to squeeze the last bit of efficiency, they can be good "rules of thumb".

You may wish to restate your goal if the above has violated any constraints.

Where there are constraints, engineering involves compromise.

If the only goal is maximum efficiency, no holds barred, we have a lot of room to redesign.

If you must stay with that particular core, operating voltage, switching frequency and circuit configuration, then we can only make minor gains in efficiency.

« Last Edit: 2011-05-21, 03:25:58 by ION »


---------------------------
Just because it has a patent application or is patented does not always mean it really works.
   
Group: Guest
thats a very old circuit there..
last year i thought i made a new circuit setup only to find that it was a hartley..
i've have built this circuit more than a hundered times.. and you dont need all the resistors in there..
i liked this video cause i half light the neon with only 120 micro amps and 1.1 volts..
i have ran this circuit upto 34 volts.. and down to 0.1 volts..
this is a very efficient circuit far better than any collector type circuit..
watch my hartley.. videos and you will see it run alot of leds in  parallel


http://www.youtube.com/user/koolerization#p/u/2/wav6mvzUdZs
   
Group: Moderator
Hero Member
*****

Posts: 2106
Professor

I have been following this thread with interest. You stated your goal in one of the posts.

With this in mind, as an old hand at switchmode design, can I ask a question and make a few recommendations ?

1) In light of your goal, what function does Rr perform? You state that it is used as a current limit. Considering this, it will certainly waste power.

2) Consider operating at higher Vin, where semiconductor Vdrop will be less significant. You will need to readjust bias.

3) Use a good quality Schottky diode rather than LED. Determine the sweet spot value for the output load resistor. The idea is to keep current through the diode low so as not to waste power there, unless you want to calculate that power and use it as part of the output consideration, even though it is unusable except as a heater.

4) With the output capacitor connected, there will be high peak currents in the diode. Best to not use it unless you need a filtered output. A small choke will limit peak current if it Cout must be used. Consider synchronous rectification for lowest loss in the rectifier device.

5) Readjust turns ratio so that power is not wasted in unnecessary base drive. There is nothing magic about a 1:1 ratio. Personally I like to use the heaviest gauge wire  I can for the main winding, while leaving enough room on the core for just enough turns of a thinner gauge to sustain oscillation. Since any decent transistor will have a current gain of at least 100, keep the ohmic losses in the main winding as low as possible.

6) If you can operate above 5 volts, consider using a FET instead of BJT, as the drive requirements will be lower.

7) If size is not a problem, use a larger core with same number of turns. The operating frequency will be lower, but this will mean less switching loss overall. Generally if you can keep the operating frequency low, core loss and transition loss will also be reduced. Judging from the waveforms in "try3" scope shot, I would hazard a guess that you have lots of switching loss from the rise and fall slopes.

Finally, I see nothing magic about the common collector configuration vs. the common emitter configuration of a standard blocking oscillator. The only slight advantage may be that base drive also appears in the inductor ramp up current, but this is two orders of magnitude less important than other lossy circuit problems.

Those are my first cut suggestions. While some of these suggestions are more applicable to power supply design in the 10's of Watts range, if you are really trying to squeeze the last bit of efficiency, they can be good "rules of thumb".

You may wish to restate your goal if the above has violated any constraints.

Where there are constraints, engineering involves compromise.

If the only goal is maximum efficiency, no holds barred, we have a lot of room to redesign.

If you must stay with that particular core, operating voltage, switching frequency and circuit configuration, then we can only make minor gains in efficiency.



"1) In light of your goal, what function does Rr perform? You state that it is used as a current limit. Considering this, it will certainly waste power."  The idea I had is to limit Pin, to increase the Pout/Pin ratio.  I have some further results that indicate this may be the case (later).



Yes, my goal is maximum efficiency, and I very much appreciate your comments, Ion.
   
Group: Moderator
Hero Member
*****

Posts: 2106
thats a very old circuit there..
last year i thought i made a new circuit setup only to find that it was a hartley..
i've have built this circuit more than a hundered times.. and you dont need all the resistors in there..
i liked this video cause i half light the neon with only 120 micro amps and 1.1 volts..
i have ran this circuit upto 34 volts.. and down to 0.1 volts..
this is a very efficient circuit far better than any collector type circuit..
watch my hartley.. videos and you will see it run alot of leds in  parallel


http://www.youtube.com/user/koolerization#p/u/2/wav6mvzUdZs

Clearly, the general blocking oscillator (also Joule Thief) circuit is quite old, no question.
I enjoyed your video, Kooler, but did not see a schematic...  Could you then provide the schematic, showing especially the LEDiode direction, and the placement of the emitter in the circuit?  Also the neon.  Would be great -- a sketch will do and will allow comparisons.
   
Group: Guest
here are some of my older schematics..
http://www.overunity.com/index.php?topic=9733.0

and i appreciate your interest in this type of osc.. so i will make you a diagram of the flyback on this circuit and share more circuits of this type..
but it may be in 10 or so hours before i get around to making any..
but i will give you alot of schematics.. even my one that uses the collector and emitter to pulse the ferrite or air core..
the only problem is that is in the 106 khz range and higher..
but one circuit is 657v for 1.5 mins on a 3300uf cap.. very low current..

robbie
   
Group: Moderator
Hero Member
*****

Posts: 2106
Hey, Robbie -- I look forward to you continued input on this one!  I'm truly learning a lot, having fun for a retired guy.  (Retired at age 57 few years ago)

OK -- I tried something I was sure would work.  I put an LED in the OPPOSITE direction across the resistor Ro at 9.7Kohms.  See schematic -- this added LED would be in place of cap Co, LED "pointing" down.   (Also, I replaced the original red LED with a yellow one to allow seeing them both, no mistaking reflections one to the other.)  So now we have TWO LED's in opposite directions -- and guess what, they both light up! Photos show the data; sorry one is particularly fuzzy w/my little HP camera.

Data:  Ch 1 (yellow LED in "D" location on schematic):  Vpp 7.04 V, Vmean 720 mV.

          Ch 2 (red LED across Ro, parallel to Ro):  Vpp 7.20 V, Vmean 640 mV, so voltages close to = over each LED, with current flowing (one could say) in both directions out of point 6.
Given that Ro is at 9.7Kohms, I'd guess not much current flows through it.... very roughly,  (few V/9.7Kohms) --  so of order, perhaps a few hundred microamps through resistor.  

So where do the LED-currents come from, sloshing through the opposing LED's, both joined to point 6 as current-source point?
Quite a puzzle, to me...
« Last Edit: 2011-05-21, 07:11:42 by PhysicsProf »
   
Group: Guest
...
So now we have TWO LED's in opposite directions -- and guess what, they both light up!
...

When reverse polarized, a LED presents a capacitance (from pF to tens pF) that allows high frequency currents to pass.
So I presume that in turn at each alternation, a LED light up with current in normal direction while the other acts as a capacitor in series.

   
Group: Moderator
Hero Member
*****

Posts: 2106
You may be right, exnihilo, about the LED's acting as caps.  Of course, the connected and opposed LED's both lighting up is a curiosity compared to the Pout/Pin ("efficiency") that I'm getting with this little circuit already using the best measurement methods I know of at this time.  If you look at post #1, you see 79mW/10mW, which is interesting.  Then post #6 shows roughly the same ratio; as I wrote:
Quote
" Different wound-toroid and circuit board from reply#1 above; replication effciency results tend to corroborate."
So yes, I think this circuit is worth pursuing a bit further, and checking out.  I may be wrong, but I think it is more than a curiosity. Thanks, Kooler, for the thread you pointed to -- lots of good material there.
 And as ION noted,
Quote
If the only goal is maximum efficiency, no holds barred, we have a lot of room to redesign.

That really is the goal, max Pout/Pin, with the best possible and redundant measurements to accompany the "redesign" development.  I thank you for joining me on this quest for understanding. OK, I'm going to call it a quest for "energy truth"...  What is going on here? is what I'd like to know.  Data-based, though; I'm not so interested in fanciful theories at this stage.
 (BTW -- Saturday, will be gone most of the day.  Look forward to comments.)
   
Group: Guest
..
   

Hero Member
*****

Posts: 1160
Kooler,

The two circuits on the right of your diagram
seem to have connection errors in that the
power source is directly attached to the transistor
collector and emitter.

Will you have a look at them again?


---------------------------
"Truth: the most deadly weapon ever discovered by humanity. Capable of destroying entire perceptual sets, cultures, and realities. Outlawed by all governments everywhere. Possession is normally punishable by death." - John Gilmore (1935- ) Author
   
Group: Guest
Kooler,

The two circuits on the right of your diagram
seem to have connection errors in that the
power source is directly attached to the transistor
collector and emitter.

Will you have a look at them again?

thanks for looking over it.. cause i am known to make mistakes..
but yes the circuits are correct..
looks weird doesn't it..
   
Group: Moderator
Hero Member
*****

Posts: 2106
here are some of my older schematics..
http://www.overunity.com/index.php?topic=9733.0

and i appreciate your interest in this type of osc.. so i will make you a diagram of the flyback on this circuit and share more circuits of this type..
but it may be in 10 or so hours before i get around to making any..
but i will give you alot of schematics.. even my one that uses the collector and emitter to pulse the ferrite or air core..
the only problem is that is in the 106 khz range and higher..
but one circuit is 657v for 1.5 mins on a 3300uf cap.. very low current..


robbie

Intriguing circuits, Robbie.  Can you point to schematic, this one: "one circuit is 657v for 1.5 mins on a 3300uf cap"?  Amazing.  Have you ever measured Pout/Pin on this?
I'd be glad to do this measurement if you will permit a replication...
PS -- I hope some will replicate the little sj1 circuit from post #1....  and help draw a proper schematic of it.

Pps --  LCR Meter for $17. delivered: http://cgi.ebay.com/LCR-RCL-INDUCTANCE-CAPACITANCE-RESISTANCE-METER-W-Leads-/260749773429?pt=LH_DefaultDomain_0&hash=item3cb5e5ca75    I ordered one so I can measure L especially, on my hand-wound toroids etc.
   
Group: Guest
...
Can someone re-do this schematic with SPICE or equiv? thx, much obliged.
...
      

Here is a rough LTspice simulation in the attached txt file. It must be renamed in ".asc" in order to be open in LTspice (the forum doesn't accept files with asc extension although LTspice asc are text files. Could it be corrected?).
It is just a base to go further.
K1 is a directive for establishing the mutual coupling coefficient of L0/Lb.

On the jpg, the green trace is the emitter voltage. The blue trace is the CSRin current.

We see that we are far from Mhz frequencies. Either the real inductances have lower values or the toroid core is saturated, phenomenon that LTspice doesn't deal with, except if we give it the model for a real toroid transformer including the saturation curve (rather tedious to make).

We have also to replace ideal NPN and diode with real ltspice models (not yet checked, but I think there is one for 2N2222 and for 1N4148) and then to see what is the input/output energy. It is unlikely that LTspice shows overunity because it is based on the Kirchoff's laws of currents. Only real experiments could show OU.

   

Group: Administrator
Hero Member
*****

Posts: 2606
It's not as complicated as it may seem...
Guys,

"asc" files have now been enabled for attachments.

Thanks,
.99
   
Group: Guest

Thanks for the asc, poynt!

   
Group: Moderator
Hero Member
*****

Posts: 2106
  Thanks for this, exnihiloest...

  In my head, I get about 100 KHz for your simulation results... which indeed is far from the 1.4 MHz observed. 

Its late, and I need to retire, but I've checked the frequency for the Lb-Cb tank circuit... which controls the transistor gating, clearly.  I get 1.37 MHz (see attached), which agrees with observed.
  I'm surprised the sim is so far off from the observed freq.
Thanks again!
   
Group: Moderator
Hero Member
*****

Posts: 2106
Thanks for the change, .99.   Any chance you'll do a replication of this circuit, and tests including Pout / Pin?

 In my 2nd post above, I noted good results with Rr = 3.1 ohms.
Quote
Ro 9.8K
Rr 3.1ohm
Rb 2K
No  Co
I(t) measured over 1ohm CSR's

There are reasons, mainly good results, that I would like to show the Rr ("return resistor") in the circuit, XNest, as I had in my schematic drawing.  Could you add that to your nicely drawn schematic kindly? It will not affect the sim much, adding 3.1 ohm, but I sometimes increase that resistance and take Pout/Pin measurements, so I would like to SHOW the presence of this Rr in the diagram, if you would pls.

One of my main concerns is that the Pout calculation in the Tek 3032 may have a DC offset in the voltage measured across the CSRout, which would result in an incorrect quantitative measurement of Pout, although qualitatively it does appear positive on both the Tek and on the ATTEN scopes, and significantly larger than MEAN Pin.  MEAN Pin is close to zero as the input current oscillates around zero.  But being QUANTITATIVE accurately is not so easy.  It is an intriguing little circuit.

I get the two opposing LED's in series to light up brightly now, with Rout removed entirely from the circuit.
   

Group: Administrator
Hero Member
*****

Posts: 2606
It's not as complicated as it may seem...
Thanks for the change, .99.   Any chance you'll do a replication of this circuit, and tests including Pout / Pin?

Professor,

I'm currently on vacation. When I return, I will likely be building the oscillator Rose is currently working on. Glen has sent me some additional MOSFET's, so I can build the full version. I'll be doing some testing on that circuit.

Hopefully the sim for your circuit can be developed to produce results closer to what you are seeing. That way you'll be able to probe anywhere in the circuit and perhaps determine where the differences are between the sim and bench unit.

Regards,
.99
   
Pages: [1] 2 3 4 5 6 7 8 9
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2017-06-27, 13:14:49