Anybody used these ferrite cores ?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,
We normally use E-shaped cores.
Toroids are difficult to wind. And would only be useful for small power smps. The bars are mostly used for antenna.
Have a look on EPCOS website. There is another manufacturer, but I am having a brain freeze.
Cheers.:xeye:
 
Ferrite toroid transformers are useful sometimes, for example in 12V (or 24V) powered high-current applications where no mains isolation is required.

Concerning this ebay offer, I would ask the seller about the manufacturer and the material of the cores (and also wether they are coated or not) before buying. Knowing only that permeability is 2000 is very little data.
 
Hi Eva,

I contacted the seller... He told that it`s russian made ferrite toroids.
Due to m=2000, it`s should to be 2000HM or 2000HH ferrite - very poor quality , & unstable materials. (I was living in ex USSR for 19 years, so I have some expirience with russian made parts)
I found better supplier at Amidon website. In addition to their own cores they are selling a lot of other brands like EPCOS, Magnetics, TDK, Fair-Rite etc. Their 77 and F materians are seems to be intresting. Their prices & shipping rates are very resonable. I ordered some toroids to try them out ...


HAVE GREAT HOLIDAYS GUYS !!!
 
Tolik said:
........I found better supplier at Amidon website....... Their 77 and F materians are seems to be intresting. Their prices & shipping rates are very resonable. I ordered some toroids to try them out ...

HAVE GREAT HOLIDAYS GUYS !!!

Tolik-

I have used the Amidon #77 Ferrite toroids with excellent results in both 12V input DC-DC and off-line AC-DC supplies. Regarding off-line, toroids should NOT be used if there is a high primary-to-secondary turns ratio (like 38 : 2 for 5V outputs). I use these numbers based on my calculations for my particular application. Of course, they are material- and frequency-dependent. Toroids are much better suited for lower ratios, like 38 : (8+8), or similar, where the secondary windings can easily cover the entire core. Also, it is a good idea to wind half the primary, tape, secondaries, tape, and then second half of primary. This will ensure good pri-sec magnetic coupling.

For powdered-iron toroids, the big yellow one in almost every AT/ATX psu is the #26 material. I have ordered these from Amidon in a variety of sizes: 50, 68, 80, 94, 106, and 130 sizes.

Steve
 
Hi Steve,

Thank you for sharing your experience.
I know about magnetic leakage losses problem. I wont to try to design
one small SMPS for my digital soldering station project, and one powerfull DC DC for car amp. For offline SMPS I ordered EA 77 500 E-type core, & for DC-DC 2 toroids made from 77 & F materials. Will see them soon.
By the way , I have some MOSFETS : IRFP460 for offline, - should be good for 100W SMPS, and IRF3205 or STP80NF06 for DC- DC.
What is your suggestion for Car amp SMPS - IRF or STP ?

sorry for my English
 
Definitely IRF. International Rectifier has extremely high quality, and I trust it implicitly. '460s, huh? These are hefty MOSFETs. I still have a rail of old MTW20N50Es, (500V 20A), the Motorola equivalent to the IRFP460. Sadly, ONSemi (Motorola's spin-off) no longer makes any MOSFETS above 200Vdss.

The IRF3205 is a very good choice for your car amp. Parallel a few of these for the DC-DC section, and you're ready to rock-n-roll. For rectifiers, I have used 4 of either the MBR20200CT (200PIV, 20A Common-Anode Schottky), or the MUR1520, or the MUR1620CT (+), and MUR1620CTR (-). Using the complimentary '1620 pair yielded a (+) rail about 600mV higher in magnitude than the (-) rail. I recommend using the MUR1620CT & -CTR only if space is at a premium, otherwise, go with four individual diodes. As for the Schottkys, I use these because of their lower forward drop. A few percentage points better in efficiency, I think.
 
I agree, IR is my favorite power MOSFET`s manufacturer.
But I still intresting about STP80NF06. If somebody tried, share your experience please.

For rectifiers, I have used 4 of either the MBR20200CT (200PIV, 20A Common-Anode Schottky), or the MUR1520, or the MUR1620CT (+), and MUR1620CTR (-). Using the complimentary '1620 pair yielded a (+) rail about 600mV higher in magnitude than the (-) rail. I recommend using the MUR1620CT & -CTR only if space is at a premium, otherwise, go with four individual diodes. As for the Schottkys, I use these because of their lower forward drop. A few percentage points better in efficiency, I think.

What about IR diodes, I mean HFA25PB60 HEXFRED or same diodes with soft recovery? HEXFRED`s from IXYS are also seems to be intresting...
Today I recived my order from Amidon - 3 cores. Packed very well to prevent damage during shipping.
I found some datasheet about amidon`s stuff. It`s 19mB .pdf file, so if anybody intresting I can to shot it by email...

HAPPY NEW YEAR !!!
 
I totally forgot about IR's diodes. Duh. Though I have never worked with them, I would have to imagine that their diode quality is right up there with their MOSFETs.

I have had no experience with the IXYS components, though I have seen them advertised in most of the well-respected trade journals, so they must be OK.

Good on theAmidon stuff. It's been a while since I've dealt with them, but all my experiences with them have been good ones. I especially like their data sheets and databook. Now that that info is available online, I think I will download it, since I'm always mis-placing my sheets and databook.

Happy New Year to all as well.................
 
I especially like their data sheets and databook. Now that that info is available online, I think I will download it, since I'm always mis-placing my sheets and databook.

I also wery liked their clearly & simple discriptions, but their website don`t contain any downloads :( I found very old datasheets that I using now. The new Amidon`s databook cost $8, & avaible by order only. I think I`ll pick up one with my future order.

I totally forgot about IR's diodes. Duh. Though I have never worked with them, I would have to imagine that their diode quality is right up there with their MOSFETs.

I found that Borbely Audio claiming that these are the best hi power diodes - http://www.borbelyaudio.com/power_supplies.asp But here is mistake in discription - their writed that these diodes are from Infeneon. Not rigte, it`s IR.

Before I started with my SMPS projects, I maked some transformer calculator in EXEL. It`s not completed yet, but take look on it please. All critics or suggestions will be welcome.
 

Attachments

  • copy of pulse transformer calculation.zip
    12.3 KB · Views: 81
sss,

this is the best site for calculations
Intresting site, but I don`t see calculation for push-pull topology as well as calculations of driver for mosfet`s / IGBT`s arrays. It`s important fo switch key elements FAST... The most common IC like TL494 or SG3525 ar able to supply up to 200-500mA only.
Example:
To open IRFL1010N we need to charge the gate up to 6V at least .
The gate charge will be around 60nCl.
If additional regulation for driver stage not used the gate of mosfet will be charged almost up to 10-11 V (at 12V car amp).
So to switch off such mosfet we need to discharge around 90nCl from gate to ground.
Now let`s imagine that our SMPS working at 100kHz with duty sycle 80% for both switching periods. Each mosfet will be open for 4mkS, closed for 6mkS. The time of switching should not to exeed 3% of time while mosfet is open. Otherwrise termal dissipaton on mosfets will increase dramaticaly. So we need to switch our mosfets within 120 nS max. Now we can to calculate current for charge / discharge gates.
Ig+= 60nCl/120nS= 0.5A
Ig-=90nCl/120ns=0.75A
In additon capacitance of gates is non linear - so currents will be higher than calculated above. And this is for 2 MOSFET`s only.


were u able to make a car amp cheaper then those available in the market in israel?

There are main 2 reasons to build your own amp:
1. It`s not for save money- but it`s for pleasure
2. Genesis or McIntosh amps are very very expencive here as well as every other things ;) I think we able to build an amp much better than amps assembled in China for price much lower than price of amps that was handcrafted in USA or England :)

btw , i speak russian too

Generally this is international forum so let`s respect other users here, but feel free to contact me by email or messenger using russian language :)
 
i always use buffers to drive the mosfets (made from solid state transistors)

I fouund better stuff on ebay-
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&rd=1&item=150184571372&ssPageName=STRK:MEWN:IT&ih=005
http://cgi.ebay.com/UC3710-6A-MOSFE...Z005QQcategoryZ109455QQtcZphotoQQcmdZViewItem
http://cgi.ebay.com/UCC27321-MOSFET...Z005QQcategoryZ109455QQtcZphotoQQcmdZViewItem

simple to use, and very powerful chips that able to drive large arays of
mosfets/IGBTs . Sometimes he making auction for items above so if you laky enouth, you can to get set of 5 chips for $3-5...
I dealed wit this seller a lot , and I must confess that he is fery realable
& frendly guy.
 
You don't need a too big SMPS for car amplifiers (particularly for class D ones). With the help of suitable gate resistors and ferrite beads (so that current becomes limited to safe values), SG3525 can drive one or two pairs (or even four) of MOSFET. Switching times may be a bit long, but in push-pull car SMPS the turn-on event should be resonant and should happen at zero Vds (so extreme speed is not required), and the turn-off event can't be too fast because the transformer radiates an EMI spike proportional to turn-off di/dt (and to leakage inductance, so toroid winding arrangement is critical for EMI).

Take a look at my class D car amplifier with built in SMPS: http://www.diyaudio.com/forums/showthread.php?s=&threadid=111566&highlight=

EDIT: The server where the images in this thread are hosted seems to be down. I'm trying to solve the problem.
 
Tolik said:
btw, witch amplifier you builded for your car ? I mean witch schematic you used ? How it sounded ?

i started with rod's project 68 (its 68 isnt it?:rolleyes: )
and then i did my own circuit for 500Wsmps and then
another my own circuit with uc3825 controller ( i dont remember but i think it was that) it also used ETD59 core :hot: for allot of power , i dont really know how much ,i disassembeled it because i want to use the core for 1kw or so home amp .
oh ,i have no car , i ride a motorcycle ,weeee :D


Eva said:
You don't need a too big SMPS for car amplifiers

Eva , my ears are bigger then yours!
:D :D :D
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.