using mosfets as rectifiers?

[Winfield Hill]

Ken Smith wrote...

I have someone else checking their records. I didn't turn up a
copy of the schematic. I think I, remember where the charge
current came from.

Remote preamp Long cable Main unit
................. +15 .....................
+15V .--------------------------.
! . GND .
/ .--------------------------.
\ . .
/ . .
\ . Signal in question .
!----------------------------------------- Reciever
!- . .
!- Q1 . .
!- . -15 .
! .--------------------------.
-15V . .
................. .....................

With Q1 on, the signal is at -15V.
With Q1 off the signal is at +15V
This works fine until we add some more info.

There is some big currents switching in the pre-amp section. Q1 turns on
just as the other thing switches off and remains on for a few mS. The
other circuit switching causes the preamp's GND to jump vs the ground in
the main unit. The combination of the cable capacitance and the receiver
circuit caused Q1's current to flow backwards. Either the bias on Q1
wasn't enough to keep the voltage below 0.7V or the current was backwards
just when Q1 was supposed to go off.

The cable could be as much as a mile long. The large signalling voltage
swings were needed to deal with the ground differences.

OK, got that. Now describe the rest of what happens, the symptoms.

Thanks,
- Win

(email: use hill_at_rowland-dot-org for now)

 

[Ken Smith]

Ken Smith wrote... [.. schematic snipped ..]

With Q1 on, the signal is at -15V.
With Q1 off the signal is at +15V
This works fine until we add some more info.

There is some big currents switching in the pre-amp section. Q1 turns on
just as the other thing switches off and remains on for a few mS. The
other circuit switching causes the preamp's GND to jump vs the ground in
the main unit. The combination of the cable capacitance and the receiver
circuit caused Q1's current to flow backwards. Either the bias on Q1
wasn't enough to keep the voltage below 0.7V or the current was backwards
just when Q1 was supposed to go off.

The cable could be as much as a mile long. The large signalling voltage
swings were needed to deal with the ground differences.

OK, got that. Now describe the rest of what happens, the symptoms.

It looked something like this. This was over 10 years ago so I'm not
certain on the details:

***.................................**********.
....*...............................*..........
....*.............................**...........
....*............................*.............
....*...........................*..............
....*..........................*...............
....*.........................*................
....*.*..*............********.................
....*.*.**......*..***.........................
....**.*.**********............................
......*.........*..............................
1 2 3 4 5 6 7


At 1, the big stuff starts to turn off
At 2, there was some crashing around as coil voltage reverses
At 3, there was a blip as we go from fixed voltage to resistive damping
At 4, the coil current has stopped and the gate drive is removed.
At 5, the MOSFET's drain voltage has risen to some part way voltage
At 6, the voltage begins a normal shaped exp. rise to +15V
At 7, the voltage is most of the way up.

The 4-6 time was several mS long (as in maybe 3)

 

[Asimov]

"Jim Adney" bravely wrote to "All" (09 Jun 04 20:30:53)
--- on the heady topic of "Re: using mosfets as rectifiers?"

JA> From: Jim Adney <[email protected]>
JA> In my limited experience, Schottkys have a lower forward drop than
JA> even Ge.

Ge diodes are about 0.3 volts Vf while I think Schottky diodes have
about 0.5 volt but a microwave type diode can be less than 0.1 volts.

A*s*i*m*o*v

.... A couple of volts below threshold.

 

[Jan Panteltje]

In my limited experience, Schottkys have a lower forward drop than
even Ge.

Perhaps depends, I looked up AA119 Ge diodde, and 250mV at a few mA.
Schottky depends a lot on the current?
Now I look at BAT85 datasheet, at 25C it is 300mV at 10 mA, a lot better
at higher temp and lower current.
Hey, maybe we could make a xtal radio with a Schottky!

JP

 

[Albert]

I don't follow this stuff much, so I didn't know that such things
existed. I still have to ask: Is 2.2V where the turn-on starts, or
where it is really fully on?

It's where it starts to turn on. However, even if the resistance was
only 100 ohms between the source and drain, it would probably be
enough to key the transmitter (keying line on a modern ham
transceiver). At best, I need to control a milliamp of output
current and worst case would be 50 ma of output current needed. So I
can use a partially conducting mosfet.

Thinking about this, I gather that you intend to run these MOSFETs
backwards, so that the "rectifier" is on when it is parallel to the
body diode and off when in the usual forward direction. Does that
work? I never considered the possibility....

No.....

I asked about mosfets as diodes. I was hoping I could bias the gate
high and the mosfet would only conduct with positive input
voltage...but it appears they are somewhat unidirectional. Since I
need to control the input to the gate when positive input voltage is
detected, I probably can't use this method as I don't have a steady
supply voltage as the unit is self powered (no external dc input
voltage).

I am however going to use a mosfet to switch my load, which is the
keying line on the radio transceiver.

So, mosfets as rectifiers is out. Mosfet as a switch to pull the
keying line low when I get enough audio voltage WILL be used.

In my limited experience, Schottkys have a lower forward drop than
even Ge.

Even is the Ge drops a slightly lower voltage, the leakeage current is
much higher, which is one reason why we use silicon. There just isn't
anything to gain by switching to Ge.

I don't know your application, but I wonder if you could transformer
couple your signal and boost it a bit, just enough to give you the
extra voltage you need to keep this job simple.

With a bipolar transistor acting as a switch, I am probably limited
severely by the transformer step up ratio. But, the mosfet needs very
little real power applied to the gate to make it switch, so looking
for a higher turns ratio audio type transformer might be the answer as
well.

I have to say that I've been reading all the responses and I've
learned alot. But, you (Jim) seem to have the clearest understanding
of my original question and my reason for seeking help here.

I wonder what the practical limits are for turns ratio for 8 ohm input
impedance? If I could get a 50 to 1 step up ratio, I would have no
problem generating enough voltage to turn my mosfet switch on! I don't
think transformers can be built with that much of a step up ratio (and
still be a relatively wide range audio frequency device).

I thank all who have commented.

Regards,

A

 

[H. Dziardziel]

I have an extremely low power application (recovering a small amount
of dc power from the audio output of a PC soundcard). I am trying to
key a transmitter when the soundcard puts out audio. It's actually a
self powered VOX, deriving all it's needed power from the audio output
of the soundcard. Right now, I'm almost succeeding, but need another
half volt or so to make the keying reliable.

I am using my laptop computer, which has only speaker out jacks. I am
feeding the speaker output into the low impedance side of an audio
transformer and taking the stepped up voltage from the secondary (8
ohm to 1K ohm impedance transformer).

Laptop headphone outputs are usually more than 8 ohms. The 8 ohm
winding transformer may be needlessly pulling down levels even
with an unloaded secondary..

snip

 

[Michael Schwingen]

It's powering the gate of a small signal mosfet. The mosfet needs
about 3 volts to turn on completely. The mosfet is hooked to ground
and the positive keying terminal on the microphone jack. It (the
keying line) is already pulled up by the electronics inside the radio,
so it needs to be pulled down to ground in order to key the
transmitter. The mosfet pulls the keying terminal to ground when it
conducts.

Can't you steal a tiny bit of current from that keying line without
activating the key input to power some low-power Opamp? That would be enough
power for the turned-off state. When you activate the key, you would need to
get power from the audio input (or provide a large enoug buffer capacitor
for the longest on-interval).

cu
Michael

 

[El Meda]

With a bipolar transistor acting as a switch, I am probably limited
severely by the transformer step up ratio. But, the mosfet needs very
little real power applied to the gate to make it switch, so looking
for a higher turns ratio audio type transformer might be the answer as
well.

You can use the same transformer that you already have, and use two
schottky diodes and two capacitors to make a voltage duplicator, then
apply that voltage to the gate of the mosfet:



|
||-+
||<-
o-----o-||-+
- | |
^ | |
|| | | |
-||-----o | |
|| - --- |
^ --- |
From transformer | | |
--------o-----o----o

created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

 

[Winfield Hill]

Albert wrote...

I wonder what the practical limits are for turns ratio for 8 ohm
input impedance? If I could get a 50 to 1 step up ratio, I would
have no problem generating enough voltage to turn my mosfet switch on!
I don't think transformers can be built with that much of a step up
ratio (and still be a relatively wide range audio frequency device).

Step-up and step-down transformers are basically the same, but
connected backwards. Up to 10,000-ohm CT to speaker (4-ohm) used
to be made for driving speakers from vacuum tubes. The impedance
ratio is the square of the turns ratio, so that'd be a 1:50 xfmr.

Radio Shack used to sell transformers similar to this. Now they
only offer the # 273-1380, rated at 1000:8 ohms or an 11:1 ratio.

Ocean State Electronics has many types of small audio transformers.
http://www.oselectronics.com/ose_p119.htm You could use two of
these, say a 1000:8 (p/n 45-714) plus a 100000 to 1000 (p/n 45-702)
to achieve an impressive overall 112:1 ratio.

Thanks,
- Win

(email: use hill_at_rowland-dot-org for now)

 

[Spehro Pefhany]

Albert wrote...

Step-up and step-down transformers are basically the same, but
connected backwards. Up to 10,000-ohm CT to speaker (4-ohm) used
to be made for driving speakers from vacuum tubes. The impedance
ratio is the square of the turns ratio, so that'd be a 1:50 xfmr.

Radio Shack used to sell transformers similar to this. Now they
only offer the # 273-1380, rated at 1000:8 ohms or an 11:1 ratio.

Ocean State Electronics has many types of small audio transformers.
http://www.oselectronics.com/ose_p119.htm You could use two of
these, say a 1000:8 (p/n 45-714) plus a 100000 to 1000 (p/n 45-702)
to achieve an impressive overall 112:1 ratio.

Thanks,
- Win

(email: use hill_at_rowland-dot-org for now)

Mouser also carries (carried?) a wide range of "import" type small
audio transformers at very reasonable prices compared to Pico et al.

Best regards,
Spehro Pefhany

 

[Jim Adney]

Ge diodes are about 0.3 volts Vf while I think Schottky diodes have
about 0.5 volt but a microwave type diode can be less than 0.1 volts.

Certainly the Vf depends on the current in all three types, but I
think that I get more like 0.2V when I just check a Schottky on my
Fluke. I've been using 1N5819s and 1N5820s to improve the efficiency
of some low voltage switching supplies that I often repair, and the
results seem consistent with this.

Isn't a microwave diode still a silicon PN junction? If so, then it is
only 0.1V by virtue of the low forward current.

It's clear that we have to be careful here, because we all know that
we can find a silicon diode that will measure 0.1V with 1A and another
silicone diode that will measure 1V with the same current. It's just a
matter of how close to the max I rating for the diode we are.

I've got some nice 10,000 A silicon hockey pucks at work. I wonder
what they show on a Fluke? ;-)

-

 

[Asimov]

"Jim Adney" bravely wrote to "All" (11 Jun 04 19:06:15)
--- on the heady topic of "Re: using mosfets as rectifiers?"

JA> From: Jim Adney <[email protected]>

JA> On Thursday, 10 Jun 2004 10:18:28 -500 "Asimov"

Ge diodes are about 0.3 volts Vf while I think Schottky diodes have
about 0.5 volt but a microwave type diode can be less than 0.1 volts.

JA> Certainly the Vf depends on the current in all three types, but I
JA> think that I get more like 0.2V when I just check a Schottky on my
JA> Fluke. I've been using 1N5819s and 1N5820s to improve the efficiency
JA> of some low voltage switching supplies that I often repair, and the
JA> results seem consistent with this.

Of course it depends on Is how the Vf curve goes. For power
rectification, say at about 3A, an Si diode will exhibit about 1V and
a similar rated Schottky about 0.5V. Now for small signal applications
this all goes out the window especially as it approaches the leakage
currents. For example a 1N914 will show between about 500K and 1M ohm
small signal resistance at Vf=0 (with signal input 10 times less than
Vt of about 25mV) but I think this is due those leakage currents.
Small signal transistor C-B junctions are often much better than that
though the capacitance tends to be larger.


JA> Isn't a microwave diode still a silicon PN junction? If so, then it is
JA> only 0.1V by virtue of the low forward current.

I think a microwave diode is designed to take advantage of quantum
effects. It works with waves and fields as the operating principle for
signals but probably like an ordinary diode for DC and low
frequencies, as they seem to have a frequency range spec. For example
odd things like the output starts to change before the input signal
actually propagates to the output. Now if I had a diode that would
have an output before I actually applied a signal, now you're talking!


JA> It's clear that we have to be careful here, because we all know that
JA> we can find a silicon diode that will measure 0.1V with 1A and another
JA> silicone diode that will measure 1V with the same current. It's just a
JA> matter of how close to the max I rating for the diode we are.

Especially with pulse currents when a rectifier supplies current to a
very large filter capacitor. For example Vf may be 1V at 3A but what
does it become with a 100A current pulse?


JA> I've got some nice 10,000 A silicon hockey pucks at work. I wonder
JA> what they show on a Fluke? ;-)

A negative reading? i.e. it supplies a measurable saturation current
just from the temperature photons in your hand? ;-)

A*s*i*m*o*v

.... Over a hundred billion electrons were used in crafting this tagline.