ultra
Member
You can get this from the core data sheet. All manufacturers have temperature listings.
Easy method of removing ferrite core
- Thread starter isa_olcer
- Start date
You can get this from the core data sheet. All manufacturers have temperature listings.
rikkitikki
New member
You will crack the core if heated to quickly due to thermal stress. It will lose its magnetc properties at the curie point (some hundres C) but it is reversible. Eventually at some 1500 C it will start to melt or sinter down.
Ferrites are produced by sintering the powder at 13-1400 C .
Iron powder toroids and the likes are composed of mellaitc powder and a resins binder
Ferrites are produced by sintering the powder at 13-1400 C .
Iron powder toroids and the likes are composed of mellaitc powder and a resins binder
Magnetic core property will be affected at certain high temperature.
rikkitikki
New member
Yes, but at such high temperatures(curiepoint) that by then any resins binding the core halves are carbonized. I realised my previous answer was incorrect, some parameters are affected when the core is heated to the curie point.
I tried the hot air method suggested earlier.
The transformer came apart very eaily.
I rewound the primary with 360 Turns of 0.2mm diam wire and and 16T of 0.3mm diam wire for the secondary.
It passes a 7Vpp square wave @ 87kHz beautifully. Output is ~250mVpp for an ESR meter.
The transformer came apart very eaily.
I rewound the primary with 360 Turns of 0.2mm diam wire and and 16T of 0.3mm diam wire for the secondary.
It passes a 7Vpp square wave @ 87kHz beautifully. Output is ~250mVpp for an ESR meter.
http://ludens.cl/Electron/esr/esr.html
I changed the opamp to high voltage version. It's all I had in stock that was cheap. I have a small stock of opa2134 but was not willing to sacrifice one to this experimental project.
It runs well at +9Vdc single supply but the output justs start to be corrupted @ 8.5Vdc.
So I set the regulator to 8.9Vdc. This means I cannot run the tester from a 9V battery. So I use the output of my adjustable lab supply set to around 11.5Vdc.
This results in a higher output voltage from the oscillator. So instead of increasing the primary turns to more than 400T, I reduced the secondary turns from 20T to 16T. It gives 250mVpp instead of 200mVpp as the square wave test signal.
The resolution of the meter near the Xc value of the capacitor under test is not good. Difficult to measure the difference between 1ohm and 1.2ohms. This makes an enormous difference to the actual ESR if the Xc were 0.9ohms.
But the meter clearly shows faulty caps and is capable of showing esr that has degraded to say 1.5times Xc for values over around 0.5ohms
I am thinking of changing the 10r in series with the DUT to 1r or 2r to improve resolution at the low ohms end and maybe a little bit more gain in the second opamp.
Xc = 1 / {2 Pi F C) in my case the F is 87kHz (11.5us from the scope screen).
The meter reading is {esr + Xc}.
This is a bigger value than esr alone that some meters can read.
I have just bought an Ebay component tester to do esr and many other component measurements. But these tools cannot check esr when the cap is in circuit.
The Ludens meter's main advantage is in circuit testing.
I changed the opamp to high voltage version. It's all I had in stock that was cheap. I have a small stock of opa2134 but was not willing to sacrifice one to this experimental project.
It runs well at +9Vdc single supply but the output justs start to be corrupted @ 8.5Vdc.
So I set the regulator to 8.9Vdc. This means I cannot run the tester from a 9V battery. So I use the output of my adjustable lab supply set to around 11.5Vdc.
This results in a higher output voltage from the oscillator. So instead of increasing the primary turns to more than 400T, I reduced the secondary turns from 20T to 16T. It gives 250mVpp instead of 200mVpp as the square wave test signal.
The resolution of the meter near the Xc value of the capacitor under test is not good. Difficult to measure the difference between 1ohm and 1.2ohms. This makes an enormous difference to the actual ESR if the Xc were 0.9ohms.
But the meter clearly shows faulty caps and is capable of showing esr that has degraded to say 1.5times Xc for values over around 0.5ohms
I am thinking of changing the 10r in series with the DUT to 1r or 2r to improve resolution at the low ohms end and maybe a little bit more gain in the second opamp.
Xc = 1 / {2 Pi F C) in my case the F is 87kHz (11.5us from the scope screen).
The meter reading is {esr + Xc}.
This is a bigger value than esr alone that some meters can read.
I have just bought an Ebay component tester to do esr and many other component measurements. But these tools cannot check esr when the cap is in circuit.
The Ludens meter's main advantage is in circuit testing.
Last edited:
Thanks AndrewT, that Luden is one smart guy.
Here is another ES meter i found, not free but it looks like it has a lot of respect.
http://shop.anatekcorp.com/index.php?
src=directory&view=products&srctype=detail&refno=2&category=Component%20Analyzers
BTW Luden uses the oven method to take cores apart.
Here is another ES meter i found, not free but it looks like it has a lot of respect.
http://shop.anatekcorp.com/index.php?
src=directory&view=products&srctype=detail&refno=2&category=Component%20Analyzers
BTW Luden uses the oven method to take cores apart.
Last edited: