rmeyn
Full Member
My primary interests are electronics, sailing, bowling, and camping
Posts: 163
|
Post by rmeyn on Apr 10, 2009 13:36:47 GMT -8
This possible to me. I've simulated a similar transformer before and you have to couple the coils correctly. The real problem will the pot core. I'll give it a try. Richard
|
|
rmeyn
Full Member
My primary interests are electronics, sailing, bowling, and camping
Posts: 163
|
Post by rmeyn on Apr 10, 2009 17:44:14 GMT -8
Gary, Iamazoid I ran this circuit on LT spice, used coils with inductances as shown and coupled them. I got a pulse output which ranged between -30V and 210V at 10k hertz. I have not played around with the other magnetic parameters and I took a guess on amperages. I did not have a 2n3055 available but found one similar. Is this output what would be expected? I can post the output results if you like. richard
|
|
|
Post by Gary Lecomte on Apr 10, 2009 19:40:49 GMT -8
Gary, Iamazoid I ran this circuit on LT spice, used coils with inductances as shown and coupled them. I got a pulse output which ranged between -30V and 210V at 10k hertz. I have not played around with the other magnetic parameters and I took a guess on amperages. I did not have a 2n3055 available but found one similar. Is this output what would be expected? I can post the output results if you like. richard I would guess the pulsed output is about 134 Volts, based on the Turns ratio. Its actually a very simple circuit.
|
|
|
Post by lamazoid on Apr 10, 2009 23:57:31 GMT -8
Gary, you are right, this is exactly my circuit. Actually, i've tried to run it also in "classic" manner, when the feedback is connected to the base throught capacitor. Practical result is almost the same, but simulation does not run with 2nd variant. Richard I dont' know what is actual voltage output, it can be different when tube strikes and when it lights up. In my design, power consumption from 12v car battery is 0.7A and it drives 8w tube almost at the same brightness as commercial driver. My goal for this simulation is to reduce Q1 heating (currently ~50 degrees C) and use smaller core for transformer if it's possible. Also i've noticed that many designs over Internet using just an AM antenna rod for transformer. I want to play with this also, but i have no clue how so simulate open core
|
|
|
Post by Gary Lecomte on Apr 11, 2009 5:50:31 GMT -8
Lamazoid, An "AM Antenna rod" is designed for high frequency signals only. Not for Power applications. So it would not be very Efficient.
But there are Many Ferrite Rods that LOOK Simular and will work good with power circuits.
However, Pot cores are More Efficient than rods. What is the Physical Size of your pot core?
Maybe just use a Smaller one?
|
|
|
Post by lamazoid on Apr 11, 2009 8:31:50 GMT -8
I've tried 18 and 14mm pot cores. It works fine with 18mm , i had even built one complete lamp with it. With 14mm core something strange happens: lamp is shining good (at the same brightness) , current consumption is also the same, but transistor gets VERY hot after few secs of operation. The number of turns is reduced, only 8 turns for both primary and feedback. Later i will post some pictures of all transformers.
|
|
|
Post by Gary Lecomte on Apr 11, 2009 13:33:48 GMT -8
I've tried 18 and 14mm pot cores. It works fine with 18mm , i had even built one complete lamp with it. With 14mm core something strange happens: lamp is shining good (at the same brightness) , current consumption is also the same, but transistor gets VERY hot after few secs of operation. The number of turns is reduced, only 8 turns for both primary and feedback. Later i will post some pictures of all transformers. Generally, You need a reasonable Inductance on the Primary winding, verses the frequency or your transistor may dissipate too much heat. Or it could be that your Frequency has gone up too high for the transistor. The Feedback winding usually isn't as critical.
|
|
|
Post by lamazoid on Apr 12, 2009 8:11:14 GMT -8
Gary, can give any general info about this oscillator? I should know at least, which way to go im my experiments. I can't find any practical information at all
|
|
|
Post by Gary Lecomte on Apr 12, 2009 12:33:00 GMT -8
Gary, can give any general info about this oscillator? I should know at least, which way to go im my experiments. I can't find any practical information at all Not knowing What Equipment you have, "IF ANY", I First suggest you look at some Inductive Reactance Charts. (Possibly in a "Radio Amateur Handbook" from your Public Library.) And Knowing the "AL Value" of your Pot core, will help you determine its inductance. If your Ferrite Data Sheet Doesn't give it, Determining AL Value is explained on my Projects page.
|
|
|
Post by lamazoid on Apr 13, 2009 1:15:26 GMT -8
I have an inductance meter, so i have no problem determining induction value for the coil or AL value for the core. My problem is oscillator itself. How to calculate its waweform (frequency and duration) , output voltage, power, how to pick up correct transistor, etc, etc.
|
|
rmeyn
Full Member
My primary interests are electronics, sailing, bowling, and camping
Posts: 163
|
Post by rmeyn on Apr 13, 2009 5:44:39 GMT -8
Iamazoid, I was able to simulate your design if I used a ideal, linear transformer as outlined. However, when I tried to simulate it with your magnetic properties and as 3 non linear inductors, I can't get it to oscillate. Were you sucessfull with LTspice? I've got a feeling I am missing something with that pot core! Richard
|
|
|
Post by lamazoid on Apr 13, 2009 6:17:39 GMT -8
Yes, it works in LTspice when configured as shown on previous page. I can email you ciruit file if needed.
|
|
rmeyn
Full Member
My primary interests are electronics, sailing, bowling, and camping
Posts: 163
|
Post by rmeyn on Apr 13, 2009 7:39:22 GMT -8
Iamazoid, Yes please, just the netlist will do. I must have an error somewhere in the attributes. Richard
|
|
|
Post by lamazoid on Apr 13, 2009 10:53:26 GMT -8
Here it is: "ExpressPCB Netlist" "SwCAD III Version 2.20p" 1 0 0 "" "" "" "Part IDs Table" "C1" "47n" "" "C2" "330µ" "" "R1" "1000" "" "R2" "6k" "" "Q1" "2N3055" "" "V1" "13.4" "" "L1" "30µ" "" "L2" "70µ" "" "L3" "22m" "" "L4" "Hc=16 Br=0.1 Bs=0.35 Lm=0.037 Lg=0.0001 A=0.000045 N=200" "" "L5" "Hc=16 Br=0.1 Bs=0.35 Lm=0.037 Lg=0.0001 A=0.000045 N=20" "" "L6" "Hc=16 Br=0.1 Bs=0.35 Lm=0.037 Lg=0.0001 A=0.000045 N=10" "" "Net Names Table" "N005" 1 "0" 5 "N001" 12 "N002" 17 "N004" 20 "N003" 23 "Net Connections Table" 1 1 1 2 1 3 2 3 1 7 2 4 1 12 2 0 2 1 2 6 2 2 2 7 2 4 2 8 2 5 3 9 2 6 2 10 2 9 2 11 2 10 2 0 3 2 1 13 3 3 1 14 3 6 1 15 3 8 1 16 3 11 1 0 4 4 1 18 4 9 1 19 4 10 1 0 5 5 1 21 5 8 2 22 5 11 2 0 6 5 2 24 6 7 1 25 6 12 1 0 schematics was modified a bit during my experiments, so i decided to make new picture to prevent any mismatch: and here is the result of simulation, showing collector voltage of Q1, voltage on load and input current:
|
|
rmeyn
Full Member
My primary interests are electronics, sailing, bowling, and camping
Posts: 163
|
Post by rmeyn on Apr 13, 2009 10:58:29 GMT -8
Thanks, Looks like you are simulating your original circuit, not the one proposed by Gary. Thanks Richard
|
|