Square wave Pulse Induction

[deemon]

Здравеите г-н Киреев,благодаря за вашия отговор -Г-н Кирееваз съм много зле с Английския език поради което мога да Ви пиша на Руски ,Немски или на Български език.Мисля,че това не е проблем но бих искал да чуя вашето мнение по този въпрос.Желя Ви хубави почивни дни. 173...

Да тут правила такие - надо на английском , иначе забанят

Написал в приват на русском .

[kt315]

i can give you direct link on same subject in Zakrzchevski's forum. if you need.
you can post in your native or russian there.

[Qiaozhi]

Здравеите г-н Киреев,благодаря за вашия отговор -Г-н Кирееваз съм много зле с Английския език поради което мога да Ви пиша на Руски ,Немски или на Български език.Мисля,че това не е проблем но бих искал да чуя вашето мнение по този въпрос.Желя Ви хубави почивни дни. 173...

Please read the forum rules -> Basic Rules of the Forums
and make your posts in English.

This is the 3rd time I have had to warn you about non-English posts, and you've only made 3 posts!

[geoschew]

Благодаря за предупреждението, което ще спазвам и публикувам на английски, използвайки Google translator.Geoshev.

[kt315]

write in this thread http://www.md4u.ru/viewtopic.php?f=8&t=6518

Interestingly, did anyone try to make a pulsed metal detector with energy recovery, that is, with the return of energy stored in the search coil - back to the power circuit? After all, there all the energy is spent on the resistor parallel to the coil, because these devices consume so much. I?ve been thinking about this topic for a while, and I?m wondering - did anyone else do this?

[Qiaozhi]

Благодаря за предупреждението, което ще спазвам и публикувам на английски, използвайки Google translator.Geoshev.

English!

Note to everyone else: Please do not reply to Non-English posts.

[ivconic]

George i bet you have prepared macro for that action?

[Altra]

This is my attempt to reproduce Deemon's amazing CCPI tx circuit.

Still a work in progress. It uses a minelab SD,GPX style DD IB coil. This proto uses manual current compensation. Deemon devised a self adjusting current tilt, which I wanted to avoid while learning.

The power transformer is the key to the circuit. Deemon was using recycled toroid cores and could not ID the material type. So I set out to determine which standard cores work. I used ferrite cores with 25mm to 29mm O.D.s. Ferrite material types tested "F", "J", "W" and metal tape type "4D". One example - 29mm J material, primary 130t #32 and 70t, 2 x #26 teflon for the secondary. The primary should have about 1 ohm resistance. On the secondary the lower the better.

Most toroids you find are the powdered iron type which will not work. Permeability between 2800u to 8000u seem to be the sweet spot. I even made a transformer from a salvaged E-core that was used in a electronic light balast. It worked but was physically too small for the heavy gauge secondary.

Attached are Deemon's drawings that I based my experiments on. Also transformer test circuit along with a proto that I am still playing with.

[Elliot]

Watching with interest. The preamp-output wave form looks problematic (including coil not being dampened properly).

[eclipse]

I don't think this circuit needs damping, take a closer look.

[Altra]

The Tx coil is undamped. The rapid reversal of the current kills the coil voltage after T = 3.14 * sqrt(LC) .The Rx coil is critically damped. Coil is induction balanced and nulled. Most of the mess you see on the pre-amp output is some odd harmonics of the half sine. You can actually sample during this period. I have not tried this because the current compensation kicks in during the peak of the half sine. You can see on the yellow trace a notch at the peak. This is about +/- 20v being added to correct the current loss between pulses.

[green]

This is my attempt to reproduce Deemon's amazing CCPI tx circuit.

Still a work in progress. It uses a minelab SD,GPX style DD IB coil. This proto uses manual current compensation. Deemon devised a self adjusting current tilt, which I wanted to avoid while learning.

The power transformer is the key to the circuit. Deemon was using recycled toroid cores and could not ID the material type. So I set out to determine which standard cores work. I used ferrite cores with 25mm to 29mm O.D.s. Ferrite material types tested "F", "J", "W" and metal tape type "4D". One example - 29mm J material, primary 130t #32 and 70t, 2 x #26 teflon for the secondary. The primary should have about 1 ohm resistance. On the secondary the lower the better.

Most toroids you find are the powdered iron type which will not work. Permeability between 2800u to 8000u seem to be the sweet spot. I even made a transformer from a salvaged E-core that was used in a electronic light balast. It worked but was physically too small for the heavy gauge secondary.

Attached are Deemon's drawings that I based my experiments on. Also transformer test circuit along with a proto that I am still playing with.

Your scope pictures show a longer delay time between target sample and ground sample than I've been using. Have you found the long delay time to be better?

[Altra]

Your scope pictures show a longer delay time between target sample and ground sample than I've been using. Have you found the long delay time to be better?

Hi Green,

Those settings were random. I was actually learning how to use the Rigol screen print feature. If set, the sample delay would have been shorter and the gb would have been much closer to the first sample. The ccpi seems to have different gb characteristics than other BPPI's. I'm not sure where to define the start of the sample delay. If you look at the digital turn on pulse, there is about a 4us delay before the current reversal begins. What is the target seeing during the different time periods? Still lots to understand.

[green]

Hi Green,

Those settings were random. I was actually learning how to use the Rigol screen print feature. If set, the sample delay would have been shorter and the gb would have been much closer to the first sample. The ccpi seems to have different gb characteristics than other BPPI's. I'm not sure where to define the start of the sample delay. If you look at the digital turn on pulse, there is about a 4us delay before the current reversal begins. What is the target seeing during the different time periods? Still lots to understand.

Hi Altra,

Don't know your Rigol scope. With my Rigol, I select Bit map and(invert)to print a white screen instead of black screen. My preference, maybe others prefer black screen?

[dbanner]

Dbanner will ring in with a comment here.
I intend w you guys are dedicated to the task at hand. Surely you are devotees to your cause.
Now there are delays which are intrinsic to any circuit which started as an aberration in one's mind.
At the microseconds level one is, without knowing, dealing with a different beast.
Think, Altra.