Diode losses...

[Appleon]

Hi guys,

I am using MAX761 to construct a regulator circuit from a solar power suuply. However, I found that the output capacitor is actually slowly discharging when no input is available. In fact, it is decreasing in voltage even with no load. I have isolated the problem to be an issue with the schottky diode 1N5817 recommended in the datasheet, the problem being that the capacitor seems to be dicharging from this diode. After I switched the diode to a diode chip, the drop in voltage across the output capacitor is reduced significantly. Can anyone tell me why this is so? The diode chip I substituted is actually a rectifier bridge DF04M, but it shouldn't be an issue. Has this drainage of charge from the capacitor got to do with the reverse characteristics of the diode? I read the datasheet for the 2 diodes but am not very sure what is mentioned in them, just that it seems that when the diode is in reverse-mode the current for the DF04M seems to be a lot less. Is this the reason why? The chip max761 is a switching regulator, so could this be reacting with the 1N5817 in generating the loss? Do note that the MAX761 in question is operating in non-bootstrapped mode. Ideas?

Thanks in advance!!

Cheers!!
Appleon

Note: I can't seem to be able to upload the datasheets, but the circuit I am building is exactly like that for non-bootstrapped mode for MAX761 in the datasheet.

 

[Roff]

At 25C, maximum reverse leakage on the 1N5817 is 1ma. On the DF04M, it's 5ua.
On the Schottky 1N5817, forward voltage at 1 amp is typically about 400mv. On the PN junction diodes such as the DF04M, Vf is closer to 1 volt. That's one of the tradeoffs in some applications - do you care more about the reverse leakage or the forward drop? For your application this is moot. Reverse recovery time on a PN junction rectifier is too slow for the switching frequency of the MAX761 (see p.10 in the datasheet). You have to use a Schottky diode, which has very fast reverse recovery time.
So why do you care about a little reverse leakage? It ia surely insignificant compared to your load current (isn't it?).

 

[Appleon]

Reverse Losses...

From the above reply, am I right to say that the losses is definitely due to the reverse losses of the schottky diode? Actually, I have a query about the frequency of the regulator. What does this frequency relate to? Is it the frequency of the switching action inside the regulator? In this case, if my diode is not capable of fast switching speeds, what will happen? More power losses? For the purpose of my charger, I was actually considering placing a load capacitor before any load is connected. I wanted to maintain the potential of this capacitor for as long as possible when the supply is off so that for short durations the device I am powering can still obtain the necessary voltage. When I placed the schottky diode across, the power dissipation action is much too significant. When I substituted the diode with the pn diode, this dissipation is a lot less significant. In this case, wouldn't a pn diode be more useful, unless there is some adverse effects if I use it? Comments?

Thank you Ron :wink:

Cheers!!
Appleon

 

[Roff]

Please answer these questions before I attempt to answer yours:
How much current does your load draw? Is it on the order of a few milliamps? What is the length of the "short duration" you expect the cap to hold charge?

 

[Appleon]

Hi,

Guess I was asking a stupid question, since my load is actually drawing at least 1A. No sense wanting to maintain the current feed even for short durations of the order of a few minutes. Thanks.

Cheers!!
Appleon