Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Better shottky diodes

Status
Not open for further replies.

Powerglide

Member level 1
Joined
Dec 7, 2019
Messages
39
Helped
4
Reputation
8
Reaction score
4
Trophy points
8
Activity points
290
I'm building a push pull dc to dc converter and when I measure the voltage on secondary of the ferrite transformer it gives me 25.5 AC output which is correct for the amount of turns on the secondary.
When I measure after the full wave rectifier , which consist of 2 mbr4045pt shotkky diodes I get a voltage reading of 16volts dc on a 160 watt load.
Is it possible to get such a high voltage drop with the shottky diodes and what other diodes can I use to improve this?
Also, is it possible to use a mosfet as a shottky diode in reverse bias with its internal diode?
 

A number of questions:
1. How are you measuring the 25.5V AC (what kind of instrument are you using)?
2. Is it an RMS measurement and if so what is the wave shape?
2. Is the 25.5V measured while the 160W load is applied?
3. How are you measuring the 16V DC?

And the answer to the final point - no, the drop will be MUCH lower than that. If you are losing 10.5V with a 10A load, it implies the diodes are dissipating over 100W of heat in them.

Brian.
 

Hi,

How did you measure this?
I doubt your measurement results.

You need to do it with a scope to see what´s happening.

****
A DVM in AC mode may show:
* recitified average
* true RMS
* estimated RMS
* or any other..
* all the above with or without DC

And it may be optimized for
* mains frequency
* pure sine waveform

Thus a DVM generally (in my opinion) is not useful to measure high frequency switching signals.


Klaus

added:
A MOSFET will be worse than a schottky. Your MBR4045PT should be O.K.
 

Yes, I measured with a dvm and while it was on the 160 watt load.
It might not be accurate but the difference stays the same percentage with a lighter load.
I haven't used the scope before to measure the voltage , as I'm not fimiliar with it, but I'll try to do it and see if I get more accurate results.
I'm having voltage drop issues and this was the most obvious to see.
 

I think that confirms your measurements are wrong.

Most DVMs lose accuracy at frequencies higher than one or two KHz and your statement that you used a ferrite transformer suggested it was much higher than that. Also as Klaus hinted, a DVM is calibrated to read a sine wave input, any deviation from that might give a wrong reading and it is quite possible your waveform is nearer a square wave. You really need an oscilloscope or a true RMS meter to take the measurement.

The other issue is that DC measurement. Without a schematic we can't tell if you have any reservoir capacitance after the rectifier, if not it means you are measuring the average of DC pulses rather than a steady voltage.

Brian.
 

Your switchmode is the cause - it is not powerful enough or it is limiting by some action of control loop - or the input source cannot supply the power required

look a bit wider with your DMM ...
 

1) Schottky diodes are inferior to PN rectifiers for high
current series resistance, given equal die area and
breakdown. This, because Schottkies as majority
carrier devices have no conductivity modulation in
forward biased operation. The modest Vf advantage
is taken up by I*R soon enough.

2) Your VAC measurement may be right but consider
that to conduct into a 16V load, you are only picking
off the crests of the sine wave. This means most of
the cycle you get nothing, and you have to get your
DC average current by much higher peak current
during the useful slivers of the cycle. Back to the I*R
problem, only the I that matters is some multiple of
what your load current says.
 

The modest Vf advantage is taken up by I*R soon enough.

Is it really true?

Say I am comparing one Schottky and a PN junction diode with similar current rating (no point comparing devices beyond their specified limits).

If I understand correctly what you are saying, the I-V graph of the Schottky diode will intersect and cross over to the right of the I-V curve for the junction diode (within their specified limit)?
 

Generally yes. But the basis of rated current cannot
be assumed identical.

Current rating may or may not track die area, it has
other constraints such as wire bond diameter, number,
package thermals etc. A diode with better Vf@If(max)
can have a higher rated current as its Pdiss will be lower.
Conversely a PN diode with same rated current may then
have a smaller die, and similar series R.


But voltage goes pretty straight to drift region thickness
and doping and area, with that, to resistance (t*rho/A).
PN diodes modulate that downward in high level injection
while Schottkies do not.

Fast recovery diodes may be a superior option but these
often trade conductivity modulation for storage time
reduction (either lifetime-killing implants like Au or Pt,
or distributed "recombination plugs" that getter minority
carriers that stumble into their local repletion region,
or some combination).

A PN rectifier with too-slow recovery will let charge
slosh back out from the load. That's also an efficiency
hit. There's a "sweet spot" involving forward loss,
reverse losses (Qrr and leakage) and the frequency
and duty cycle of all that.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top