IXYS High Voltage current regulator
sudeep...@gmail.com
response & how about IXCY10M90S how about the heating Issue ?
Fred Bloggs
> Haws any one tried IXYS current regulator ? IXCY10M45, wats the
> response & how about IXCY10M90S how about the heating Issue ?
>
IIRC IXYS or whoever has at least one app note on the subject even a
simian subspecies can understand...
Winfield
> Has anyone tried IXYS current regulators? IXCY10M45, what's
> the response & how about IXCY10M90S. What about heating issues?
These devices are interesting to me, in that they must use
depletion-mode MOSFETs as the pass element. This gives them
a free low-voltage power source to run a reference and opamp,
to control the MOSFET's gate, and create a constant current.
But, one must ask, why not just use a depletion-mode MOSFET,
such as offered by Supertex, to create the constant current,
and avoid the integrated reference and opamp. One reason
might be to obtain an improved current-source accuracy, but
the IXCY' parts are spec'd with a huge nearly 50% part-part
variation, plus another 15% tempco variation. It seems the
extra integrated elements aren't providing much improvement
over ordinary HV depletion-mode MOSFETs, especially when one
consideres how much better-than-their-specs the depletion-
mode MOSFETs are in practice. Maybe the wide-voltage-range
current constancy is where the smart parts excel. But the
fact that one can't buy these parts from distributors could
be an additional troublesome factor.
John Larkin
wrote:
Interesting idea: use a Supertex depletion fet to power an opamp and a
reference in its own source circuit, which in turn actively regulates
the current. That would make a precision, high-voltage, 2-terminal
current regulator.
How's this for starters?
|
|
|
|
d
+-------g
| s
| |
| |
| ---+---
| | |
| | ldo |-------+
| | | |
| |_______| R
| | |
| | |
+--------+-----------+-----------
John
John E. Perry
> ...
>
> Interesting idea: use a Supertex depletion fet to power an opamp and a
> reference in its own source circuit, which in turn actively regulates
> the current. That would make a precision, high-voltage, 2-terminal
> current regulator.
>
> How's this for starters?
>
>
> |
> |
> |
> |
> d
> +-------g
> | s
> | |
> | |
> | ---+---
> | | |
> | | ldo |-------+
> | | | |
> | |_______| R
> | | |
> | | |
> +--------+-----------+-----------
>
>
> John
>
>
John, if you connect the gate to the ldo's output, you get a common
circuit from the application notes that has super-high output resistance
and accuracy commensurate with the quality and accuracy of the ldo and
resistor you use.
John Perry
Winfield
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Sun, 19 Aug 2007 18:59:16 -0700, Winfield <winfieldh...@yahoo.com>
You typically don't get a very high -Vgs for the
depletion-mode MOSFETs, whereas you need 1.2 volts
for the LDO reference, plus the maximum dropout
voltage, so I'd move the FET's gate to the LDO's
output pin, like this:
high-voltage |
input source |
d
g ----,
s |
| |
___|___ | Io = 50uA + 1.2/R
| | |
| LDO |----+--,
|_______| |
| R
| |
'-----------+-----------
There are troublesome issues: the required BFC
output-stabilizing cap (not nice for a current
source), the minimum load current spec, the
rapidly-disappearing -Vgs at higher programmed
currents, the ugly power dissipation with high
input voltages, etc., but it's a cute idea.
John Larkin
wrote:
Nice touch... gives the ldo more room to work and gets closer to Idss.
John
John Larkin
wrote:
Sure. But if this stuff were easy, anybody could do it!
John
paul @removeppinyot.removecom Paul P
pencil sketch on alt.binaries.pictures.radio please? or directly to me.
Perhaps I can relate the differences for the future.
I am interested in this solution since IXYS emailed me saying their current
regulators are out of production. I use these to replace 50A1 nitrogen
filled regulator ballast tubes in Transoceanics.
Thanks,
Paul P.
"Winfield" <winfie...@yahoo.com> wrote in message
news:1187611796.8...@d55g2000hsg.googlegroups.com...
JeffM
>I must apologize. I can not discern the ASCII schematics.
ASCII schematics are done in a fixed-width font.
Ways to handle this are:
1) Set your newsreader for a monospaced font like Courier.
2) Cut & paste the text to something like Notepad.
3) Go to Google groups.
http://groups.google.com/group/sci.electronics.design
Append &fwc=1 to the URL
http://groups.google.com/group/sci.electronics.design/browse_frm/thread/6a60502cb6b925a5&fwc=1
or click on **More options** then **Show original**.
http://groups.google.com/group/sci.electronics.design/msg/376252b5f22f8490?dmode=source
.
.
We don't much appreciate top-posters in this group either.
Trim and bottom post here.
Winfield Hill
> Winfield wrote
>> John Larkin wrote:
> I am interested in this solution since IXYS emailed me
> saying their current regulators are out of production.
> I use these to replace 50A1 nitrogen filled regulator
> ballast tubes in Transoceanics.
I see the Amperite ballasts generally work with more
than 6V across them (can you confirm for the 50A1),
so an ordinary MOSFET could be used with an ordinary
LM317 3-terminal regulator to make a ballast-tube
substitute, sort of like this:
positive ----+--/\/\--,
| |
d |
n-chan g ----+
MOSFET s |
| \_|_ 7.5V
___|___ /_\
| | |
| LM317 |----+--, Io = 50uA + 1.2/R
|_______| |
| R
| |
'-----------+----- negative
Winfield
>
> I see the Amperite ballasts generally work with more
> than 6V across them (can you confirm for the 50A1),
> so an ordinary MOSFET could be used with an ordinary
> LM317 3-terminal regulator to make a ballast-tube
> substitute, sort of like this:
>
> positive ----+--/\/\--,
> | |
> d |
> n-chan g ----+
> MOSFET s |
> | \_|_ 7.5V
> ___|___ /_\
> | | |
> | LM317 |----+--, Io = 50uA + 1.2/R
> |_______| |
> | R
> | |
> '-----------+----- negative
Time to put stop playing around with theories, and
actually make one of these high-voltage medium-power
linear current-source beasts to see how they work.
Paul says a 50A1 works at 50mA, so I used R = 25 ohms.
I used an IRF640, which is a large-die 400-volt MOSFET,
with Rth-JC = 1C/W, which is good for power-handling
capability. I chose a 100k gate resistor, insuring
more than 10uA of zener current for Vin-Vout of over
10 volts (note that zener diodes rated near 5.5 to
8 volts maintain their breakdown voltages down to
amazingly-low currents).
positive ----+--/\/\--, 100k
| |
IRF640 d |
n-chan g ----+
MOSFET s |
| \_|_ 7.5V
___|___ /_\ 1n5236
| | |
| LM317 |----+--, Io = 50uA + 1.25/R
|_______| |
| Rs 24.9 1%
| |
'-----------+----- negative
This setup started conducting with an Io above 20mA
for Vin-Vout voltages above about 5V. This increased
to roughly the 50mA design current for Vin-Vout > 6V.
I was testing with a 500-ohm load resistor, and found
that I had to bypass the load with a 0.1uF capacitor,
to prevent a small (1%) erratic relaxation oscillation
with a roughly 300kHz rep rate. Further increase in
the input voltage (Vin-Vout from 10 to 50V) caused
the output current to be stiffly pegged at 50mA.
At 50V I noticed a little bit of output disturbance,
and by 75V there was some RF oscillation underway,
and the output current dropped to 46mA.
The propensity of high-voltage MOSFETs to oscillate
when used in the linear mode above 50 to 100V is well
known. I'll have to report back back later if I find
a solution to this problem (or should I say, when).
John Larkin
wrote:
The 300 KHz thing may be the zener itself; some do that at low
current. You might try bypassing the zener, then adding a 47 ohm or so
series gate resistor.
John
Fred Bartoli
Even better. I've had some work for which I had to do some tests.
Just checked a BZX85C 10V. At room temperature it roughly maintains its
voltage down to 200pA (yep, that's pA) and starts falling badly at 100pA.
Here are some figures:
Iz Vz
0.2nA 9.25V
0.5nA 9.73V
1nA 9.96V
10nA 10.27V
100nA 10.30V
1uA 10.31V
10uA 10.31V
100uA 10.31V
1m 10.33V
--
Thanks,
Fred.
Fred Bloggs
Those numbers are unbelievable, that's one incredible zener.
Fred Bloggs
Paul P wrote:
> I must apologize. I can not discern the ASCII schematics. Could you post a
> pencil sketch on alt.binaries.pictures.radio please? or directly to me.
> Perhaps I can relate the differences for the future.
>
> I am interested in this solution since IXYS emailed me saying their current
> regulators are out of production. I use these to replace 50A1 nitrogen
> filled regulator ballast tubes in Transoceanics.
>
How many do you contemplate using? Digikey is left holding the bag on
19,506 IXCP10M45S at $1.76/ea. Why don't you make a lifetime buy of
quantity 3. As I understand the Transoceanics thing, the 50A1 can be
replaced by a fixed power resistor most of the time, the tube was added
on to the design later to provide regulation of a small low current bias
voltage across installations and is not needed for any fixed
installation line supply.
Fred Bartoli
Oops, I got the reference wrong. It was a BZX55C10V (Philips).
From the tests I did some time ago, this seems to be quite general. I
only tested 500mW zeners though :-)
Here's another one (BZX55C 15V), randomly picked from my drawers:
Iz Vz
0.5nA 13.69V
1nA 13.71V
10nA 13.73V
100nA 13.76V
1uA 13.82V
10uA 13.89V
100uA 13.83V (not a typo)
1m 13.92V
--
Thanks,
Fred.
paul @removeppinyot.removecom Paul P
"Fred Bloggs" <nos...@nospam.com> wrote in message
news:46D4123...@nospam.com...
I have been looking at the IXCP10M45. What throws me it the "10" meaning
10ma. I perused the spec sheets but could not ferret out if the chip is
designed to handle the 50 ma plus of the application. I know some others
use this chip in this application (Transoceanic). That leads me to a few
conclusions:
1) I am not interpreting the spec sheets properly and I don't fully
understand the implementation that I have seen.
2) Using a more common and available product for the application. That is
what brought me here.
If the IXCP10M45 is appropriate I would make a bulk (not 19k) purchase and
make the tube sub. As for why keep this ballast in place? It is desired of
most collectors and hobbyists to modify the chassis as minimally as
possible. I recommend to others to store the 50A1 and use a sub. That way
the 50A1 keep on the shelf and not burn out. Then it can be said, "I have an
original restored Transoceanic" by the owner.
Can you tell if the IXCP chip can handle the 50 ma at 120v rms mains (1/2
wave rectified)?
Here is the application (find the 50A1 on the incoming line cord):
http://www.transoceanic.nostalgiaair.org/600/t-600_data.pdf
I regret that I am not up on current component technologies. It is just
where life takes you. So the help of this group's members is greatly
appreciated.
Paul P.
Winfield Hill
To calm the MOSFET I added a 620-ohm gate resistor, and
this worked fine. The output current stays at exactly
50mA over the voltage-drop range of 10 to 80 volts. At
80V the MOSFET is dissipating 4W, which is close to the
limit with the small clip-on heatsink I've been using.
However, above 86 volts the output current drops by 3%
and even more at higher voltages. An oscillation can
be seen at the MOSFET-source and LM317-input node.
It's about 1V at 2.5MHz, which is apparently enough
to cause the LM317 reference voltage to drop a bit.
Adding a 10nF cap across the LM317 stopped this new
oscillation. Now the observed output current doesn't
change all the way up to a 200V drop, which I tried
for only a very short time, due to an anticipated
MOSFET overheating above 80 volts.
Here's the final quiet seemingly-robust circuit:
100k
positive ----+---/\/\---, 50mA Current Source
| | with a 5 to 200-volt
IRF640 d 620 | compliance range, if
n-chan g --/\/\-+ the MOSFET heatsink
MOSFET s | is adequate.
_______| \_|_
| ___|___ /_\ 1n5236 7.5V
_|_ | | |
--- | LM317 |------+ Io = 50uA + 1.25/R
0.01 | |_______| |
|_______| Rs 24.9 1%
| |
'----------+--+--- negative
_|_
0.1uF ---
|
gnd
It does lose its appeal as the parts count goes up.
Fred Bloggs
The first number on your tubes indicates filament voltage, they are all
in series: 3V4, 1U4, 1L6, 1U4, and 1U5. This totals to 9V ( where "1" is
understood to mean 1.5V and "3" for 3.0V, which is inline with various
web page descriptions of the filament circuit operation voltage. For
best longevity with only a slight degradation of receiver sensitivity,
most users keep the total filament drop in the 7.2-7.5V or about 80%,
and this gels with original manufacturer filament life data. The 3V4 is
rated absolute maximum filament of 100mA with the two halfs in parallel,
the Transoceanic puts them in series, so 50mA would be a maximum limit
for the Transoceanic circuit. The "50" in 50A1 almost certainly means
50mA regulation. The "A" means A-series. Looking at Amperite AB-51, the
first page shows a graph of the typical A-series thresholding at 10V and
regulating up to 70V before the current deviates outside the +/-1%
regulation window, with a fairly steep rise. The same operating voltage
vs life applies to these devices too, so it is clear the reason they are
all blown out is because Zenith was running them at too high a voltage.
The T600 chassis schematic shows the 50A1 in series with 700 ohms to the
filament string. This puts 105-700x0.05-9=60V across the ballast or 85%
at 117VAC. At the more usual 125 VAC this drop maybe more like 68V,
uncomfortably close to the maximum, and shortened life, say 2000 hours
vs 10,000.
In addition to replacing the 50A1, it would be a good idea to mount a 1W
7.5V zener across the entire tube filament string. This diode will
survive quite nicely and is cheap compared to replacing those tubes.
So now all you need is a 50mA constant current, not necessary for
spectacular regulation, with operating voltage up to 100V. Because the
regulator is on the output filter side of the rectifier, it will not
require any transient protection. Expected power dissipation will be
something like (115-35)x0.05=4W, which means use a 10W clip-on heatsink
for maybe 27oC temperature rise on the outside. The IXCP10M45 can be
operated like so, derived from figures 2 and 4 of the IXCP10M45
datasheet Don't think about using this circuit without the 7.5V 1W
zener, breadboard it and wire the output into B(-)Bus as an initial
test, verifying that the voltage across the 51 ohm resistor in around
2.5V, let it cook a bit under the heat, only then connect the output
into the filaments:
View in a fixed-width font such as
Courier.
.
.
.
.
.
. from R27 and C41A
. |
. |
. |
. ------
. | A |
. | |
. .--|G | IXCP10M45
. | | |
. | | K |
. | ------
. | |
. | [51]
. | |
. '------+
. |
. |
. |
. to R33 and C41B
.
.
.
.
.
Best of luck to you, I know that radio, it's a good one.
Fred Bloggs
Congrats on your success, but it just seems so kludgey:-) Doesn't
Supertex have anything suitable...
Michael A. Terrell
>
> The first number on your tube's indicates filament voltage.
Only for MOST receiving tubes, for consumer electronics. Some, like
the 12AT/AX/AU7 had a center tapped filament for 6.3 volts OR 12.6
Volts. Some rectifiers had a 6.3 volt tap, and could be used at either
voltage. The two, three or four digit numbers on industrial tubes have
absolutely nothing to do with their filament voltage. Even more
confusing is that some consumer tubes had an industrial counterpart with
a different filament voltage, like the 6KD6 at 6.3 volts, and the 8950
12.6 volt industrial counterpart.
--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.
Michael A. Terrell
Central Florida
Fred Bloggs
Michael A. Terrell wrote:
> Fred Bloggs wrote:
>
>>The first number on your tube's indicates filament voltage.
>
>
>
> Only for MOST receiving tubes, for consumer electronics. Some, like
> the 12AT/AX/AU7 had a center tapped filament for 6.3 volts OR 12.6
> Volts. Some rectifiers had a 6.3 volt tap, and could be used at either
> voltage. The two, three or four digit numbers on industrial tubes have
> absolutely nothing to do with their filament voltage. Even more
> confusing is that some consumer tubes had an industrial counterpart with
> a different filament voltage, like the 6KD6 at 6.3 volts, and the 8950
> 12.6 volt industrial counterpart.
>
>
>
The original standard governing the tube numbers instituted by the Radio
Manufacturers Association or whoever designated the filament voltage
like I said they did, and more importantly it applies to the tubes in
question. When in doubt consult the datasheet.
Fred Bloggs
Michael A. Terrell wrote:
> Fred Bloggs wrote:
>
>>The first number on your tube's indicates filament voltage.
>
>
There was no apostrophe on "tubes" in my original post. If you want to
play that game...
Michael A. Terrell wrote:
>
>
> Ownly for MOS reseeving toobs, for consoomer electroniks. Sum, lyke
> the 12AT/AX/AU7 had a sentrr taped filiament for 6.3 voltz OR 12.6
> Voltz. Sum retifyers had a 6.3 volz tasp, and kud be yused at ether
> voltge. The too, tree or fohr didgit numers onn industral toobs hav
> absolootly knothing to doo with ther filiament voltage. Evening mohr
> confoosin is dat sum consoomer toobs had an industral cowntertop wit
> a difrent filiament voltge, lik the 6KD6 at 6.3 voltz, and the 8950
> 12.6 volt industral cowntertop.
Michael A. Terrell
The first audio and radio tubes made didn't follow any standard
numbering system, other than some manufacturers adding a 1 or 2 as a
prefix to make it appear they weren't duplicating someone else's
design. The RMA came years later. There was also RETMA, EIA and
several other "Standards" groups.
John Larkin
wrote:
>
> Here's the final quiet seemingly-robust circuit:
>
> 100k
> positive ----+---/\/\---, 50mA Current Source
> | | with a 5 to 200-volt
> IRF640 d 620 | compliance range, if
> n-chan g --/\/\-+ the MOSFET heatsink
> MOSFET s | is adequate.
> _______| \_|_
> | ___|___ /_\ 1n5236 7.5V
> _|_ | | |
> --- | LM317 |------+ Io = 50uA + 1.25/R
> 0.01 | |_______| |
> |_______| Rs 24.9 1%
> | |
> '----------+--+--- negative
> _|_
> 0.1uF ---
> |
> gnd
>
> It does lose its appeal as the parts count goes up.
I wonder if bootstrapping the zener from the reg output reduces
stability.
John
Winfield
> On Tue, 28 Aug 2007, Winfield Hill wrote:
>
>> Here's the final quiet seemingly-robust circuit:
>
>> 100k
>> positive ----+---/\/\---, 50mA Current Source
>> | | with a 5 to 200-volt
>> IRF640 d 620 | compliance range, if
>> n-chan g --/\/\-+ the MOSFET heatsink
>> MOSFET s | is adequate.
>> _______| \_|_
>> | ___|___ /_\ 1n5236 7.5V
>> _|_ | | |
>> --- | LM317 |------+ Io = 50uA + 1.25/R
>> 0.01 | |_______| |
>> |_______| Rs 24.9 1%
>> | |
>> '----------+--+--- negative
>> _|_
>> 0.1uF ---
>> |
>> gnd
>
>> It does lose its appeal as the parts count goes up.
>
> I wonder if bootstrapping the zener from the reg output
> reduces stability.
I'm happy to try it, with a 9 or 10-volt zener, but
I'd expect no effect at all, as long as the LM317
has enough voltage to operate.
Winfield
Thanks for the kind words, Fred, but it was a bit of a
disappointment seeing what it took to tame the MOSFET
solution several of us have thought about in the past.
Indeed, using one of the IXYS depletion-mode MOSFET-based
IXC current regulators, or even a depletion-mode MOSFET
all by itself, should be far more simple than the NMOS
+ LM317 kludge. Although, as you pointed out, Digikey
still has 19,506 of the IXCP10M45S regulators in stock
(down to 19,476 after I ordered 20 and someone else 10
more), it's a bit unsettling to create new designs with
discontinued parts.
One alternate is Supertex DN2540N5 depletion-mode MOSFET
in a power TO-220 package, in production, and stocked by
Mouser, http://www.supertex.com/pdf/datasheets/DN2540.pdf
They don't show a very detailed Id vs Vgs plot, but I put
some of these on the bench to get the data. One point:
unlike the sloppy JFETs of old, these modern MOSFETs seem
to have much less Vgs spread from part to part.
At low "sub-threshold" currents the dn2540's drain current
rises exponentially, just like a BJT transistor, but with
half the slope, 130mV/decade compared to 60mV. -2.75 volts
gives about 100uA. Above 1mA the rising-current plot tails
off, and we get to our 50mA goal at -2.15 volts. A 43-ohm
current-setting resistor would do the trick.
positive ------, 50mA high-voltage current source
|
d dn2540n5
,-- g
| s --/\/\--+---- negative
| 43 |
'---------------'
In taking these measurements, I was surprised to find the
'2540 really wanted to go into RF oscillation in the linear
mode, even at rather low currents like 5mA and low voltages
like 2.5 volts, so I was forced to solder a 0.1uF Vgs cap to
the part. Likewise, the self-bias circuit above oscillated
so badly above 25 volts that my DVM made faulty readings.
V I
--- ----
2 38.0mA
3 50.0
4 50.8
5 50.9
10 51.2
20 52.0
25 oops!
Adding a 1nF cap across the resistor seemed to stop this, but
I ran out of time last night to take more measurements. One
concern, at high voltages and high power dissipation with the
junction temperature rising, will the Vgs change be enough to
render the simple dn2540 50mA current source unusable?
We can hope the ixcp10m45s will shine in this regard, if its
internal voltage reference, about which we know little, is
any good. But there's one thing on the IXYS datasheet that
gives me pause, http://ixdev.ixys.com/DataSheet/98704.pdf
and that's figure 4, showing Vgs changing from -1.0 to -3.5V
as a function of current. Ahem, did I say Vgs? In other
IXYS literature on the IXC regulators they show an internal
voltage reference and an error amplifier feeding the MOSFET
gate, are we to conclude this "reference" changes by a huge
2.5 volts? It may be the IXYS part will perform no better
than Supertex dn2540 mosfet. We'll see.
Fred Bloggs
All the MOSFET structures have so many voltage dependent parameters I
doubt a single device regulator will be much good over a 10:1 VDS range.
The trick for both circuits is probably to lock the current control
MOSFET Vds with a second device which absorbs the Vds variations. I
don't know why a simple enhancement node FET w/431 combo like this will
not work:
View in a fixed-width font such as Courier.
.
.
.
.
. +
. |
. .------+
. | |
. / \ |
. | |1mA| |
. v \ / |
. | d
. +--- g
. | s
. --- |
. / \-----+
. '431 --- |
. | [Rcs]
. | |
. | |
. '------+ Vref
. | Io= ---- + 1mA
. | Rcs
. (-)
.
.
.
Winfield
> Winfield wrote:
>> Fred Bloggs wrote:
>>> Winfield Hill wrote:
>
>>>> It does lose its appeal as the parts count goes up.
>
>>> Congrats on your success, but it just seems so kludgey
>>> :-) Doesn't Supertex have anything suitable...
>
>> Thanks for the kind words, Fred, but it was a bit of a
>> disappointment seeing what it took to tame the MOSFET
>> solution several of us have thought about in the past.
>
>> Indeed, using one of the IXYS depletion-mode MOSFET-based
>> IXC current regulators, or even a depletion-mode MOSFET
>> all by itself, should be far more simple than the NMOS
>> + LM317 kludge. Although, as you pointed out, Digikey
>> still has 19,506 of the IXCP10M45S regulators in stock
>> (down to 19,476 after I ordered 20 and someone else 10
>> more), it's a bit unsettling to create new designs with
>> discontinued parts.
>
>> One alternate is Supertex DN2540N5 depletion-mode MOSFET
>> in a power TO-220 package, in production, and stocked by
>> and that's figure 4, showing Vgs changing from -1.0 to -3.5V
>> as a function of current. Ahem, did I say Vgs? In other
>> IXYS literature on the IXC regulators they show an internal
>> voltage reference and an error amplifier feeding the MOSFET
>> gate, are we to conclude this "reference" changes by a huge
>> 2.5 volts? It may be the IXYS part will perform no better
>> than Supertex dn2540 mosfet. We'll see.
>
> All the MOSFET structures have so many voltage dependent parameters
> I doubt a single device regulator will be much good over a 10:1 VDS
> range. The trick for both circuits is probably to lock the current
> control MOSFET Vds with a second device which absorbs the Vds
> variations. I don't know why a simple enhancement node FET w/431
> combo like this will not work:
> View in a fixed-width font such as Courier.
> .
> . |
> . .------+
> . | |
> . / \ |
> . | |1mA| |
> . v \ / |
> . | d
> . +--- g
> . | s
> . --- |
> . / \-----+
> . '431 --- |
> . | [Rcs]
> . | |
> . | |
> . '------+ Vref
> . | Io= ---- + 1mA
> . | Rcs
> . (-)
Thanks, Fred, I've been enjoying this conversation. Actually,
as I see it "Vds variations" are vanishingly small for these
MOSFETs, even over a 100:1 ratio, in the sub-threshold region.
Part-to-part Vgs variations may be an issue (although so far
my measurements show other wise), or Vgs vs. temp may be an
issue (the datasheets indicate otherwise, but measurements are
certainly in order). g_m could also be considered an issue,
but it's rather high in the 50mA region for these MOSFETs.
As for your suggestion of trying a TL431 variant, that appeals
to me; I will try it. Let's hope in the end that the required
parasitic components don't kill the idea. We do still have to
deal with a high-voltage MOSFET's propensity to oscillate.
paul @removeppinyot.removecom Paul P
I sent you an email this afternoon and recieved an error message. I just
resent it. Please let me know if you got it.
My email is a little squirly once in a while.
Paul P.
Fred Bloggs
Winfield wrote:
> .... We do still have to
> deal with a high-voltage MOSFET's propensity to oscillate.
>
Maybe a ferrite bead in the drain and a largeish capacitor between G-D...
MooseFET
Chances are, you need to put the loss on the source or gate to have a
much effect.
>From experience this oscillates:
60V-300V
------+----------+------
! !
\ !
/ 100K !
\ !
! !!- TO-220
+----+--!!- N Channel
! ! !!-
! ! !
20V /-/ --- +----- Vout
^ --- !
! ! ---
! ! --- 0.1u 2 places
! ! !
GND GND GND
Removing the 0.1 in parallel with the zener made the oscillation
stop. Putting a bead on the drain didn't. Putting the bead on the
source or gate did. I think the combination of the highish impedance
of the drain and the capacitance from the tab to the heatsink
prevented the bead on the drain from working.
Winfield Hill
Sorry, didn't see anything.
Winfield Hill
Zener diodes have a fair amount of capacitance, which is
sometimes enough to cause trouble. One FET manufacturer
recommends a small resistor in series with the gate. A
ferrite bead should do the same thing.
Fred Bloggs
Walt Jung has written a detailed lab report on all of these circuits:
http://waltjung.org/PDFs/Sources_101_P2.pdf
Winfield
Yes indeed, Walt's two audioXpress articles are masterpieces
of low-voltage and high-voltage current-source exploration,
as they might be used in audio amplifiers. I get that little
magazine and saw his article when it arrived, but didn't give
it the attention it deserved.
Walt also suggests reading through the work of John Broskie,
on the www.tubecad.com website. That tube crowd really goes
at it.
Yesterday I received a shipment of 20 IXYS depletion-mode
mosfets, the IXCP 10m45s in a TO-220 package.
When my tests went above 35 volts I experienced the same
oscillation problem as I had with a Supertex dn2450. Walt
features both of these parts in his article (apparently he
hasn't found any other TO-220 depletion-mode power MOSFETs
available right now either). I see he uses a 100-ohm gate
damping resistor (fig 13C), so I'll try that. I wonder if
he added that after experiencing RF oscillation, or simply
from long experience with linear power MOSFET circuits.
Walt also ignores IXYS' naming of the 10m45s pins: K and A,
instead calling them D and S, thereby treating this part as
just another MOSFET, rather than a power IC with a built-in
voltage reference and an error amplifier driving the gate,
as shown in the IXYS block diagrams. The 10m45s datasheet
plots stop at 1mA, but I plan to explore at low currents;
a quick look last night down to 100uA showed the Id-vs-Vgs
plot continuing as one would expect for an ordinary MOSFET.
John Larkin
wrote:
I always include mosfet gate resistors, even on switchmode fets. It's
good insurance against the roughly 80% probability the fet will
oscillate, and it also lets you slow down switching edges if they get
too radical.
I always use base resistors on emitter followers, too, although
emitter followers are rare beasts these days.
GaAs fets, grounded source, seem to be stable, probably because of
their impressively low Cdg.
John
Fred Bloggs
I did a cursory Google Patents search on patents assigned to IXYS to try
to get to the bottom of the 10M45 and found nothing obvious unless it's
US5629536 and/or US5751025 both High Voltage Current Limiter and Method
for Making, but they're both just cross-sectional doping profiles and
from the 1997/1998 time frame. Nothing earlier seems to match, but then
again I was brain dead at the time.
Winfield Hill
Well, the time frame is about right (product introduced
in 2000), and IXYS did say "The temperature stability
is based on a threshold compensation technique and uses
IXYS' most recently developed high voltage process".
But were 5629536 and 5751025 (originally for Motorola)
later assigned to IXYS?
Winfield Hill
> Fred Bloggs wrote:
>
>> these circuits:http://waltjung.org/PDFs/Sources_101_P2.pdf
>
> mosfets, the IXCP 10m45s in a TO-220 package.
>
> When my tests went above 35 volts I experienced the same
> oscillation problem as I had with a Supertex dn2450. Walt
> features both of these parts in his article (apparently he
> hasn't found any other TO-220 depletion-mode power MOSFETs
> available right now either). I see he uses a 100-ohm gate
> damping resistor (fig 13C), so I'll try that. I wonder if
> he added that after experiencing RF oscillation, or simply
> from long experience with linear power MOSFET circuits.
Adding a 100-ohm gate resistor stabilizes the 10m45s
up to 120V (running at "10mA" with a 287-ohm resistor).
From 120 to 200V it oscillated now and then at about
10-to-15MHz. This was using my Fluke meter as a load.
With a non-inductive 15-ohms load resistor it's stable
from RF oscillation to 260V where I stopped testing.
It was also stable with drain voltage, only changing
from 11.3 - 11.4mA from 10 - 260V. But I wonder, is
it a bit persnickety for use with a sensitive radio?
Fred Bloggs
In the T600 chassis, the 10M45 will be shunted to common by a 60uF at
(+) and 20uF at (-), C41A and C41B. It is starting to sound like the
oscillation is a result of a field effect adding additional poles into
the loop around G-S.
Winfield Hill
Actually, my guess is it wouldn't oscillate in the
radio application.
Winfield
>
> One alternate is Supertex DN2540N5 depletion-mode MOSFET
> in a power TO-220 package, in production, and stocked by
> They don't show a very detailed Id vs Vgs plot, but I put
> some of these on the bench to get the data. One point:
> unlike the sloppy JFETs of old, these modern MOSFETs seem
> to have much less Vgs spread from part to part.
>
> At low "sub-threshold" currents the dn2540's drain current
> rises exponentially, just like a BJT transistor, but with
> half the slope, 130mV/decade compared to 60mV. -2.75 volts
> gives about 100uA. Above 1mA the rising-current plot tails
> off, and we get to our 50mA goal at -2.15 volts. A 43-ohm
> current-setting resistor would do the trick.
>
> positive ------, 50mA high-voltage current source
> |
> d dn2540n5
> ,-- g
> | s --/\/\--+---- negative
> | 43 |
> '---------------'
>
> In taking these measurements, I was surprised to find the
> '2540 really wanted to go into RF oscillation in the linear
> mode, even at rather low currents like 5mA and low voltages
> like 2.5 volts, so I was forced to solder a 0.1uF Vgs cap to
> the part. Likewise, the self-bias circuit above oscillated
> so badly above 25 volts that my DVM made faulty readings.
It's likely a better approach would be simply adding
a gate resistor.
positive ------, 50mA high-voltage current source
|
d
,-- g
| s --/\/\--+---- negative
| 100 43 |
'---/\/\--------'
> We can hope the ixcp10m45s will shine in this regard ...
> http://ixdev.ixys.com/DataSheet/98704.pdf
The ixcp10m45s I measured has a substantially different
Id-vs-Vgs plot than the dn2540n5, with a strong tailing-
off of the drain current (and reduced g_m). This is a
surprise, because the 10m45 apparently has a much larger
die than the 2540, as is evident from its nice low 3.1
C/W thermal resistance, vs 8.3 C/W for the dn2540. On
this basis alone the 10m45s would be preferred.
positive ------, 50mA high-voltage current source
|
d ixcp10m45s
,-- g
| s --/\/\--+---- negative
| 100 34 |
'---/\/\--------'
I measured a nice stable 50mA current into a 15-ohm load
(dropping 7.5V, similar to the string of tubes), over a
voltage-drop range of 5 to 260 volts for the ixcp10m45s.
There was no sign of oscillation with the 15-ohm load.
(As reported, there was oscillation at high voltages if
loaded only with a 1-ohm multimeter 200mA current-range.)
If one of these setups was used for a 90-volt drop, the
FET would need to dissipate 4.5 watts without junction
overheating, which would require a serious heatsink
plus good airflow. It might be wise to add a 1k-ohm
10W power resistor in series to drop half the voltage.
Fred Bloggs
Thanks for taking the time to make all those measurements, Win.
Winfield
>
> Thanks for taking the time to make all those measurements, Win.
You're very welcome, Fred, the pleasure was mine.
I'm sorry they dribbled out slowly over so many
days, but I didn't have much time to spare for
them. I've still got a few more badly-needed
measurements to wrap up the scene.
paul @removeppinyot.removecom Paul P
Have you seen this 50A1 substtute on eBay. This seller also offeres a 1L6
replacment.
Paul P.
eBay numner#
110164235153
Winfield
> Gents,
>
> Have you seen this 50A1 substtute on eBay.
> This seller also offeres a 1L6 replacment.
>
Ah, I thought you knew about those, or had one.
I purchased one last week, and it should arrive
anytime now. I'll let you know how it's made.
Fred Bloggs
I ran across more than few descriptions of that product in my search. It
is based on the 10M45 and users say it works well. The seller has done a
really good job with these and he has a website describing the design
somewhere.
Winfield
> Paul P wrote:
>> Gents,
>
>> Have you seen this 50A1 substtute on eBay. This seller also
>
> I ran across more than few descriptions of that product in my
> search. It is based on the 10M45 and users say it works well.
> The seller has done a really good job with these and he has a
> website describing the design somewhere.
There may be several devices in the marketplace. The eBay
fellow, eea101, has great looking stuff, and also he tells
me he's a fan of AoE, so that's a good sign too!
I received it today and -- drumroll -- it's based on TI's
high-voltage LM317, the TL783. It has a bridge rectifier,
for ac operation, plus a 470-ohm 2W resistor, plus a TL783
and a 25-ohm current-setting resistor. I couldn't decipher
the resistor's color coding (I suppose it said 24.9 ohms,
but there was no red band), but it measured 25.0 ohms for
a 50mA programmed current. Along with 470R, this means
25 to 150 volts of voltage-dropping operating capability.
eea101 uses the TO-220 package part, well suited to the
serious heatsink he uses. But I note DigiKey has dropped
the TO-220 version from their stocking list, as has Newark
(why do they do that?), but Mouser has them, $1.60 ea.
paul @removeppinyot.removecom Paul P
"Winfield" <winfie...@yahoo.com> wrote in message
news:1188906578.7...@w3g2000hsg.googlegroups.com...
I have seen the eBay offering before. I have been using
http://www.dialcover.com/tubes.html one for sale, last entry. It is based
on the IXYS part. But he has been out of stock for a while.
Paul P.
Winfield
REMOVEcom> wrote:
> "Winfield" <winfieldh...@yahoo.com> wrote in message
What is he, a glass blower? Anyway, did you see my
post late last night, in answer to Fred, wherein we
learn that eea101's eBay item uses TI's TL783 chip,
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=110166475264
http://focus.ti.com/docs/prod/folders/print/tl783.html
TO-220 in stock at Mouser, p/n 595-TL783CKCSE3, $1.60
Michael A. Terrell
http://www.dialcover.com is the website of an old man nasmed Bill
Turner who used to make replacement plastic dial covers for antique
radios, along with other repruction parts. He has had multiple strokes
over the past few years. You can find him on:
news:rec.antiques.radio+phono newgroup.
--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.
Michael A. Terrell
Central Florida
Winfield
> Winfield wrote:
>> Paul P wrote:
>
>>> I have seen the eBay offering before. I have been
>>> sale, last entry. It is based on the IXYS part.
>>> But he has been out of stock for a while.
>
>> What is he, a glass blower? Anyway, did you see my
>> post late last night, in answer to Fred, wherein we
>> learn that eea101's eBay item uses TI's TL783 chip,
>> http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=110166475264
>> http://focus.ti.com/docs/prod/folders/print/tl783.html
>> TO-220 in stock at Mouser, p/n 595-TL783CKCSE3, $1.60
>
> Bill Turner who used to make replacement plastic dial covers
> has had multiple strokes over the past few years. You can
That's a great link, does he still sell and ship stuff?
I'd like to place a nice order.
paul @removeppinyot.removecom Paul P
>> http://www.dialcover.com is the website of an old man named
>> Bill Turner who used to make replacement plastic dial covers
>> for antique radios, along with other reproduction parts. He
>> has had multiple strokes over the past few years. You can
>> find him on: news:rec.antiques.radio+phono newsgroup.
>
> That's a great link, does he still sell and ship stuff?
> I'd like to place a nice order.
>
I have been calling him before placing an order. Like Mike mentioned there
are some things that are out of stock due (I believe) to his recent stroke.
Shipping is included in his prices.
Paul P.