TINKERERS_V1a_ TX

[Tinkerer]

Here is the TINKERERS_V1a_TX.

It is somewhat different to the usual PI TX.
It is different because it is a discriminating PI.
This version is optimized for a DD coil.
The damping is different and then there is the single loop compensation circuit that makes it possible to do minor adjustments of the balance of the DD coil, after the coil has been potted and even in the field, to compensate for heavy magnetic mineralization (positive bias), or salt (negative bias).

The complete circuit has been test run and is giving good results.

However, there are improvements that can be made.

Porkluvr is working on an improved power supply.

Any suggestions for improvement are welcome.

Tinkerer

[6666]

R12 shows as 100k on the gate
this seems too high
if you are trying to switch off the fet quickly, by draining the charge off the gate
that value should be under 10 k

why is it so high ?

[Tinkerer]

R12 shows as 100k on the gate
this seems too high
if you are trying to switch off the fet quickly, by draining the charge off the gate
that value should be under 10 k

why is it so high ?

6666, thanks for the feedback.

Your observation is very valid.

The TC4427 is a Mosfet driver that switches ON as well as OFF very fast. You might want to check the datasheet.

The 10 Ohm resistor reduces the noise of the switching.

So why the 100k resistor?

I work in an environment with very high noise. To avoid spurious switching caused by the noise on the gate of the Mosfet, I have made it a habit to always add a resistor from the gate to ground.
But is it needed? I should check if the Mosfet driver is open to ground, when not active.

Is 100k the ideal size for that? A small resistor will change the switching speed, but maybe 47K would be better.

Tinkerer

[ZED]

G`day Tinkerer

Looking at the schema for the TC4427 it looks like the internal fet (lower part of the totum pole CCT) pulls the gate to ground,so in this instance i think it would be safe not to use a pull down resistor on the gate of the transmit fet as i see it serving no usefull purpose.

Zed

[ZED]

R10,R24,R25,C?....interesting Damping CCT,...have you tryed it to see if it works ?
What made you decide on a 1uf cap?
How long does it take for the cap to discharge ?

Zed

[Tinkerer]

G`day Tinkerer

Looking at the schema for the TC4427 it looks like the internal fet (lower part of the totum pole CCT) pulls the gate to ground,so in this instance i think it would be safe not to use a pull down resistor on the gate of the transmit fet as i see it serving no usefull purpose.

Zed

Hi Zed,

thanks for the feedback. I agree on the resistor.

Tinkerer

[Tinkerer]

R10,R24,R25,C?....interesting Damping CCT,...have you tryed it to see if it works ?
What made you decide on a 1uf cap?
How long does it take for the cap to discharge ?

Zed

Hmmm, could you tell me which CCT you are referring to? There are so many posts and my Internet connection is slow. If it has a cap in the damping, don't try it with a ML detector. My designs are quite different, so different rules apply.
All circuits that I have posted here have been extensively tried. However, just like people make typos, I make mistakes in the schematics.

Besides mistakes, very often there are several better ways to do what I have tried. I much appreciate suggestions and observations.

It helps me a lot when people point out errors, mistakes and possible improvements.

I am trying out a new circuit design software at present. I just run the "Check Project" and have managed to come out with over 200 errors and warnings on a design with 123 pins. But in the end I will get it working.

Ah, I just scrolled up and there is the CCT in the first post. OK, the main purpose of the 1uF 400V polyester cap is to prevent the Flyback from avalanching the IRF 731 Mosfet that I used there. The CCT shows an IRF740, because that is what the EAGLE library had. Nearly the same. The combination of resistors and CAP depends on the amount of power to be discharged and is quite critical.
All the best

Tinkerer

[ZED]

Yep thats the CCT
With my own P.I. project i am endevouring to relize the same result,namely to stop the transmit fet from avalanching,most of
the heat on the transmit fet comes from the avalanch diode and stopping it from conducting during the flyback keeps things
nice and cool with higher coil currents.The method i use is a variation of the m/lab method as used in the 2000 but this method
just transfers the heat to another component,i would much prefer to use no diodes in the transmit cct including the diode in the
transmit fet.
Your little damping CCT has me interested,would you mind if i copied it to experiment with please,my P.I. limits the flyback to
around 180 to 200 volts so i figured i will have to play with it a bit to get it right.

Zed

[Tinkerer]

Yep thats the CCT
With my own P.I. project i am endevouring to relize the same result,namely to stop the transmit fet from avalanching,most of
the heat on the transmit fet comes from the avalanch diode and stopping it from conducting during the flyback keeps things
nice and cool with higher coil currents.The method i use is a variation of the m/lab method as used in the 2000 but this method
just transfers the heat to another component,i would much prefer to use no diodes in the transmit cct including the diode in the
transmit fet.
Your little damping CCT has me interested,would you mind if i copied it to experiment with please,my P.I. limits the flyback to
around 180 to 200 volts so i figured i will have to play with it a bit to get it right.

Zed

You are welcome to use the CCT.

Every diode I tried so far, produces some switching noise, so I try to avoid them too.

Since you are interested in controlling the Flyback,I am going to open a new thread for the "TWO TAU" method.

It does amazing things. We will discuss it on the forums, so if anybody tries to patent it they will get egg in the face.

Tinkerer

[Aziz]

... We will discuss it on the forums, so if anybody tries to patent it they will get egg in the face.
Tinkerer

I want to throw too. I have some bad eggs specially made for this purpose.

Aziz

[Aziz]

Energy Balance Calculations Necessary.

Hi all,

this topic gets interesting and there is an energy balance calculation necessary. I would like to support you with basic models and calculations for a simple PI switcher.

Aziz

[Tinkerer]

Hi all,

this topic gets interesting and there is an energy balance calculation necessary. I would like to support you with basic models and calculations for a simple PI switcher.

Aziz

Hi Aziz,

I am glad you are offering your knowledge.

I am not good with calculations and simulations. My circuits are from the breadboard. They work but need a lot of improvement in efficiency and quality.

Tinkerer

[Aziz]

Hi Tinkerer,

Hi Aziz,

I am glad you are offering your knowledge.

I am not good with calculations and simulations. My circuits are from the breadboard. They work but need a lot of improvement in efficiency and quality.

Tinkerer

should not be a problem soon. I did work overnight and very hard. I have new marvelous formulas and design guides to get the maximum efficiency and more.

Aziz