Toroid selection

[dr pepper]

Another couple of things you can do:

1) twist the primary and secondary together to improve coupling and stray capacitance.

2) use a low Vce sat transistor, when the transistor is on even a small voltage across it has an impact on efficiency, a good source for one of those would be one of these disposable cameras that runs of a single AA.

 

[MrAl]

Another couple of things you can do:

1) twist the primary and secondary together to improve coupling and stray capacitance.

2) use a low Vce sat transistor, when the transistor is on even a small voltage across it has an impact on efficiency, a good source for one of those would be one of these disposable cameras that runs of a single AA.

Hi,

Interesting that you should say that (#1 above). I had created an electrostatic transformer back in the 1990's for VHF and it worked without a core. I didnt have the proper equipment to test it thoroughly but it gave me better reception so i used it.

The only thing i fear here is that the electrostatic coupling will interfere with the electromagnetic coupling. It probably has to be one or the other not both, but I dont have any data on that kind of melding of principles and i dont like to make guesses with power electronics. In the past i have used one or the other but never both. Another electrostatic adventure was with a transmitter transformer which seemed to work pretty well too, again with no magnetics. That application was in the 90MHz range where electrostatic coupling is everything is coupled to everything else
My doubts about the melding of the two come from thinking that if we have low impedance due to electrostatic coupling then we must have high impedance for magnetic coupling, and vice versa, unless we know how to adjust both to ideal levels, which may produce some desirable effects or some undesirable effects.

 

[tomizett]

I don't think Dr Pepper was suggesting anything other than a conventional transformer - twisting the wires together simply ensures they are very closely coupled and the maximum amount of flux from the primary cuts the secondary.
It's ideal when you have a 1:1 turns ratio, but rather harder when the turns ratio is large.

 

[dr pepper]

i agree with you both.
however your spot on tommi.
i was and am assuning the jt has equal turns per winding that seems to be the norm even rhough maybe not the best.

 

[tomizett]

Raj:
I don't know if you will find this useful, but whenever I work with magnetics I use this spreadsheet to work out simple things like inductance and saturation current from the core data given in the manufacturer's datasheet. I'm still a bit of a novice in magnetics (and not too good at mathematics), and having a spreadsheet made up means I don't have to refer back to textbooks and get out a pencil to derive equations each time I want to wind a choke.

The spreadsheet is an ODS file, which is the OpenOffice format - I think MS Excell will open this but if not I think I can export it in XLS format. Because this forum will not allow files with certain extensions to be uploaded, I have re-named it .TXT. In order to open it, you will need to re-name it "inductors.ODS".

The mathematics behind it are simple and easy to find on the internet, this is just something I find makes life easier.

 

[Tony Stewart]

Found the inverter design: https://tinypic.com/view.php?pic=2yvm73s&s=5#.V0NE3L5V0uM

spec

Hello,

I have seen numerous Youtube videos on super high voltage Joule Thief projects. They are using small one inch ferrite toroids and from single 1.5 volt AA battery power a 3-5 watt 110-220 volt cree LED brightly consuming merely 200 mA of current from the battery! The following link will help you understand it better:

https://www.instructables.com/id/Make-a-SUPER-Joule-Thief-Light/

However, all such similar links do not specify the specific inductance of the Toroid being used or the Permeability. All they suggest is to use high permeability ferrite toroid! The authors also do not provide the answer.

Now, I want to replicate such a project at low cost. For that purpose, I have procured the following Toroids from Element14 (India):

• Manufacturer: FAIR-RITE; Ferrite Toroid: 25.4mm x 15.5mm x 12.7mm; Core Material Grade: 77; AL (nH) : 2700 ±25% (I have procured)
• Manufacturer: FERROXCUBE; TN25x15x10; Core Material: 3E25; AL: 5.62µH (I have procured)
• Manufacturer: EPCOS; Toroid; 25.3 × 14.8 × 10.0; Core Material Grade: T37; AL (nH): 6970 ±25% (yet to procure based on your advice)

As, per the above mentioned link with the "Instructible", which suggests there should be 1 volt per turn induced in the secondary of the JT, I am wondering what should be the correct toroid selected for this purpose? The toroids I have already purchased do not give me such result.

Therefore, I will be more than happy if you can guide me accordingly on this.

Thank you,
With best regards,
Raj Mukherji

this is one bad way to power a 3W LED. 9Vdc >~120Vac@40mA > *~9Vdc @330mA or 4.8W supplied but <3W out

The easier way is 9V direct to 3 series 1W (ea) LEDs, but better to use LiPo with DC PWM dimmer / timer

 

[spec]

this is one bad way to power a 3W LED. 9Vdc >~120Vac@40mA > *~9Vdc @330mA or 4.8W supplied but <3W out

The easier way is 9V direct to 3 series 1W (ea) LEDs, but better to use LiPo with DC PWM dimmer / timer

Couldn't agree more, but the OP wants to experiment.

spec

 

[dr pepper]

Hi Tommi,
I'd like to see that in .xls if you wouldnt mind.
My excel 2013 doesnt open that file.

 

[MrAl]

I don't think Dr Pepper was suggesting anything other than a conventional transformer - twisting the wires together simply ensures they are very closely coupled and the maximum amount of flux from the primary cuts the secondary.
It's ideal when you have a 1:1 turns ratio, but rather harder when the turns ratio is large.

Hi,

The word "simply" is a word that implies that something is easily understood, yet it really isnt in the case at hand. What i was suggesting was that if we improve the coupling that way then there is nothing left to excite the core. In a circuit that would be similar to reducing one resistor while keeping the other one the same value, so the lower value would now hog more of the current. And yes, if the windings are not the same length it's not going to be too easy to do

There's no way to know for sure unless we find some theory on this, based on real physics. Even winding one with and one without isnt a good enough test because one may be working better for a different reason other than what we think it might be. A good example was that when a frog's legs are cut off it no longer responds to commands to jump, so the conclusion was that the frog went deaf

We can of course look into this more.

Here is the circuit i was talking about previously. It's a boost circuit and was incorporated into a Brinkmann flashlight several years back. It is not particular to that one product only though but is a somewhat generic design.

 

[spec]

Along with the other winding improvement already suggested, Litz wire could be used for the windings, especially the secondary. http://www.osco.uk.com/products/cab...Ut6p6_IA6i_izdhXu5yueQ2tAd-kHHsc6AaAowG8P8HAQ. This will reduce the skin effect and thus reduce power losses.

As the turns ratio of a transformer increases, the efficiency tends to decrease for practical reasons, so one approach would be to possibly use fewer secondary turns and then voltage double using a couple of capacitors and a couple of diodes. Most mains LEDs will also work off DC. In fact, they normally have a rectifier on the input.

A Royer inverter would make better use of the core, and thus transfer more power from the battery. A core with a square BH curve would be best. If you would like to experiment with this approach just say, and I will do a circuit.

spec

 

[spec]

Raj,

You are talking about 100V output from the inverter, but then you mention a 240V 3W LED which is a bit inconsistent.

I suspect that the demonstrations on the net are misleading because they show a LED being illuminated- the eye has a logarithmic response to light- as the test of output power rather than measuring the actual output currents and voltages.

Just some outline calculations:

Required output power: 3W
Thus input power is: 3W * 1.3 (70% efficiency) = 3.9W
The battery voltage is 1.5V (off load) Thus average input current is, 3.9W/1.5V = 2.6A
Thus the peak transistor collector current is 2 (current saw tooth factor= 50%) * 2 (50% conduction)= 5.2A.
Assume that the transistor has an hfe of 20, iB will thus be 260 mA.

These outline calculations give you an idea of the job the transistor needs to do. So not only is the transformer design critical but the choice of transistor and its drive is also critical. This is the reason why I suggested changing from a TIP31 to TIP35, but even a TIP35 is not ideal.

A MOSFET would be ideal if it were not for the gate drive voltage- a BJT turns on around 600mV. You can get extremly low gate drive NMOSFETs, and zero gate drive MOSFETs but their other characteristics then tend to be unsuitable. Even depletion mode MOSFETs are available.

spec

 

[RAJ KUMAR MUKHERJI]

Hello MrAl,

Thanks a lot for your valuable input. I had this part of your suggestion in mind while winding the toroid. However, it somehow slipped off ....Anyways, I will give this a try with my other 3E25 toroid after undoing the present winding.

Now, I have some good news for Spec!! Yesterday, I was just testing the type 77 ferrite toroid with my JT. Sometime back, I had wound the primary on top of the secondary of this toroid with some 4-5 turns using a hookup wire (may be 24 or 25 SWG) and 14-15 turns with 30 SWG and 300+ turns for the secondary with 30 SWG. To my surprise the 3W led bulb lit up brightly off one 1.2 volt AA NiMH cell. Then I went forward and added a BD139 in parallel with TIP41C. It further improved the brightness! The light ran steadily for 3.5 hours plus before it dimmed and stopped finally. I then realized that the battery I was using was not fully charged. So, it was certainly an impressive progress and I was glad with the results. The hint from Spec did prove useful in this case.

I am now going to get a 0-500 mA analog DC ammeter to measure the current draw from the battery. I do not understand why it goes out of scale when I try to measure the DC current with my DMM using the 10A scale.

Before, I finish this reply, one question for you all...spec had advised to use TIP35x transistor for better performance. However, please let me know your views on using the 2N3055 (metallic one) transistor.

Thank you again for all your time.
Kind regards,
Raj

 

[spec]

Hello MrAl,

Thanks a lot for your valuable input. I had this part of your suggestion in mind while winding the toroid. However, it somehow slipped off ....Anyways, I will give this a try with my other 3E25 toroid after undoing the present winding.

Now, I have some good news for Spec!! Yesterday, I was just testing the type 77 ferrite toroid with my JT. Sometime back, I had wound the primary on top of the secondary of this toroid with some 4-5 turns using a hookup wire (may be 24 or 25 SWG) and 14-15 turns with 30 SWG and 300+ turns for the secondary with 30 SWG. To my surprise the 3W led bulb lit up brightly off one 1.2 volt AA NiMH cell. Then I went forward and added a BD139 in parallel with TIP41C. It further improved the brightness! The light ran steadily for 3.5 hours plus before it dimmed and stopped finally. I then realized that the battery I was using was not fully charged. So, it was certainly an impressive progress and I was glad with the results. The hint from Spec did prove useful in this case.

I am now going to get a 0-500 mA analog DC ammeter to measure the current draw from the battery. I do not understand why it goes out of scale when I try to measure the DC current with my DMM using the 10A scale.

Before, I finish this reply, one question for you all...spec had advised to use TIP35x transistor for better performance. However, please let me know your views on using the 2N3055 (metallic one) transistor.

Thank you again for all your time.
Kind regards,
Raj

Hy Raj,

Glad you have made some progress.

I am afraid to say that a 2N3055 is probably the worst power transistor you could use: Vbe, Vsat, hfe and a loose specification. The 2N3055 has two endearing characteristics: it is rugged and it is cheap. The TIP35x would be a much better choice. If you like, I will find you the best transistor for this job.

spec

 

[tomizett]

The word "simply" is a word that implies that something is easily understood

Agreed... I should have learned by now not to use words like "simple" and "easy" so readily!

Here is the spreadsheet again in Excell 97/2000/XP format.

 

[MrAl]

Agreed... I should have learned by now not to use words like "simple" and "easy" so readily!

Here is the spreadsheet again in Excell 97/2000/XP format.

Hi,

Sorry, i dont have any way to read that file, but if you could post it in a pure text file i'd be happy to take a look. Thanks.

 

[MrAl]

Hello MrAl,

Thanks a lot for your valuable input. I had this part of your suggestion in mind while winding the toroid. However, it somehow slipped off ....Anyways, I will give this a try with my other 3E25 toroid after undoing the present winding.

Now, I have some good news for Spec!! Yesterday, I was just testing the type 77 ferrite toroid with my JT. Sometime back, I had wound the primary on top of the secondary of this toroid with some 4-5 turns using a hookup wire (may be 24 or 25 SWG) and 14-15 turns with 30 SWG and 300+ turns for the secondary with 30 SWG. To my surprise the 3W led bulb lit up brightly off one 1.2 volt AA NiMH cell. Then I went forward and added a BD139 in parallel with TIP41C. It further improved the brightness! The light ran steadily for 3.5 hours plus before it dimmed and stopped finally. I then realized that the battery I was using was not fully charged. So, it was certainly an impressive progress and I was glad with the results. The hint from Spec did prove useful in this case.

I am now going to get a 0-500 mA analog DC ammeter to measure the current draw from the battery. I do not understand why it goes out of scale when I try to measure the DC current with my DMM using the 10A scale.

Before, I finish this reply, one question for you all...spec had advised to use TIP35x transistor for better performance. However, please let me know your views on using the 2N3055 (metallic one) transistor.

Thank you again for all your time.
Kind regards,
Raj

Hi,

Hey that sounds good.

Just to quote the input/output current with a few different efficiencies assuming 1.5v input and a 1.5 watt load:
100 percent: 150v out, 10ma out, requires 1.5v in, 1 amp in.
75 percent: 150v out, 10ma out, requires 1.5v in, 1.33 amps in.
50 percent: 150v out, 10ma out, requires 1.5v in, 2 amps in.
25 percent: 150v out, 10ma out, requires 1.5v in, 4 amps in.

So you can estimate your efficiency when you measure the input and output voltages and currents.

 

[spec]

I am now going to get a 0-500 mA analog DC ammeter to measure the current draw from the battery. I do not understand why it goes out of scale when I try to measure the DC current with my DMM using the 10A scale.

The current drawn from the battery will be a ramp going from about 100ma (at a guess) ramping up to around 5A. When the transistor turns off some current will flow back into the battery.

Very roughly speaking the current ramp will last about 50% of the waveform period. I would expect the frequency to be around 7KHz.

The reason why you may be getting some high reading is that your DMM is getting upset by this current waveform.

You can try putting a large capacitor across the terminals of the DMM, 1mF upwards (1000uF), to see if that gives a more realistic reading. An analog moving coil ammeter may also give erroneous readings with the current waveform.

spec

 

[RAJ KUMAR MUKHERJI]

Hi MrAl,

I will try your suggestion today/tomorrow with my 3E25 toroid - winding secondary over the primary. However, I have a question....does the efficiency/inductance depend on the 'tension' with which every turn of a coil is wound? I mean will the inductance vary between a tightly wound core VS a less tightly wound coil keeping all other parameters unchanged?

Regards,
Raj

 

[tomizett]

I think that the tighter you can wind the core the better. Less air gap will improve the coupling between the two windings, and between each winding and the core.

 

[RAJ KUMAR MUKHERJI]

Hello,

I tried a coil with the secondary over the primary yesterday. However, did not find much difference in the light output with my other design - primary over secondary. With both the two types of winding, my 3W 220V LED bulb remained on for about 4+ hours fairly with the same brightness from a 1.2V 2100mAh NiMH battery. The light was bright enough to be used a night lamp.

While tweaking the wires from the toroid on my breadboard, I noticed that at some stage (could not determine which particular point it was exactly) the light became quite bright with my combination of TIP31 and BD139 in parallel. I am yet to buy the TIP35C.

My prime target is to get moderate light output from the 3w led bulb for 3-4 hours running on 1.5 volt general purpose AA cell. I have ordered another about 1" (OD) ferrite toroid with permeability of 10,000 and AL of 12 uH per turn square and core material: "W" for this purpose. I will make it with Sec over Pri and try to use TIP35C and see the results - if there is any improvement. However, please advise if the 14T of the primary should be wound just beside the 4T of the primary or on top of it before I wind the 350T secondary on top of these two primary coils.

There is another observation: I attached a 1000uF/25V cap across the battery input of the JT circuit and tried to measure the input current with my DMM with 10A scale DC. It still went out of range!! Any more ideas on this will be greatly appreciated.

Thanks and regards,
Raj