Any beefy coil driver chips?

[Joerg]

[...]

"It's not cast in concrete yet but something like 24V industry rail with
tons of noise on there, coil about an ohm so it can ramp up to 3A or so
in a fraction of a msec, then must be held there for a msec or so, then
ramped down to hold (roughly 1/5th) in 100usec. All this must fit onto
the real estate of a postage stamp and dissipate next to nothing, as in
<500mW. If it doesn't it slowly cooks out the relay box, as has happened
on the heat pump controller we had in Europe."

... and "No supply rail". Of course 24V industrial means it should take
40V peaks.
Inductance?

"...held there for a msec or so"?? How many "so's"??

Not determined yet, could be 3, 4, 5msec. In a circuit like this it
should not matter, I just make that time selectable with a component
such as R or C. It is something that will most likely have to be changed
later. Usually the allowed tolerance is large. If it had to be under 10%
there would always be the old CD4060 which nowadays even comes in TSSOP.

Inductance??

Around 10mH or a bit higher, also not cast in concrete yet. However, in
the end this doesn't matter much, it's just the principle of operation.
No problem to do it in all discretes but the layouter will have a cow.
Why camn't anyone make a L295 in CMOS? With all that automotive stuff
nowadays there should be a market for a 60V rated device. Maybe one of
your clients could make a profit there, although since Lopez they may
not like automotive much anymore.

Just fixed our wireless doorbell extender. Major "de-crudding", a D cell
had decided to take a significant pee. Why is it that out of a set of
leak-proof alkalines one has to leak?

 

[Joerg]

Jamie wrote:

[...]

Back in the hay days, we used to have dual coil solenoids, one as
the start coil that was in series with a contact on the relay.. It
simply open when pulled in, allowing only the low current coil to
stay energized. But I'm sure this doesn't fit your application.

Not enough room. If that was possible they wouldn't need me


I think I can do it with one NAND schmitt trigger section, 2 Rs, 2 Cs,
driving a mosfet.

Seriously, it ain't quite that easy. Just one of the nastier effects:
Such lines can hang, rattle, sizzle, and then you get serious
undervoltage coming in for a while, FET tries its darndest to hang in
there ... *PHUT* ... red lights flash, sirens wail ...

I'm kind of curious as to the type of solenoid coil you're using?

The timing spec's you are spitting out seems a little fast for a
conventional coil? I would think the induction in the coil wouldn't
allow the response times you are dealing with?

Care to tell us the H value of the coil?

Not much is known at this point but the trick with this kind of stuff is
always the same: Design a coil that is rated for a small fraction of the
applied voltage. Then apply the full brunt until it reaches nominal
current, throttle back, leave it there until the relay contact has
leaned into its desired position, ramp down to hold. That's what many
chips can do but they are usually too wimpy, lossy, and large. CMOS
versions are more enticing but usually too low in rated voltage.

 

[Joerg]

Jim Thompson a écrit :

Jim Thompson wrote:
[...]

"It's not cast in concrete yet but something like 24V industry rail with
tons of noise on there, coil about an ohm so it can ramp up to 3A or so
in a fraction of a msec, then must be held there for a msec or so, then
ramped down to hold (roughly 1/5th) in 100usec. All this must fit onto
the real estate of a postage stamp and dissipate next to nothing, as in
<500mW. If it doesn't it slowly cooks out the relay box, as has happened
on the heat pump controller we had in Europe."

... and "No supply rail". Of course 24V industrial means it should take
40V peaks.
Inductance?

"...held there for a msec or so"?? How many "so's"??

Not determined yet, could be 3, 4, 5msec. In a circuit like this it
should not matter, I just make that time selectable with a component
such as R or C. It is something that will most likely have to be changed
later. Usually the allowed tolerance is large. If it had to be under 10%
there would always be the old CD4060 which nowadays even comes in TSSOP.
Inductance??

...Jim Thompson

voltage*rise_time/current

Of course. I'm just trying to get Joerg to be concise. As usual he's
as vague and side-stepping as Larkin ;-)

Hey, the specs _are_ vague at this time. That's normal when
electro-mechanical parts haven't been designed yet. Other things I am
not at liberty to say. All I was looking for were hints about chips I
may have missed, I did not intend to go into the actual design because I
know how to do that stuff

 

[Jim Thompson]

Jim Thompson wrote:
[...]

"It's not cast in concrete yet but something like 24V industry rail with
tons of noise on there, coil about an ohm so it can ramp up to 3A or so
in a fraction of a msec, then must be held there for a msec or so, then
ramped down to hold (roughly 1/5th) in 100usec. All this must fit onto
the real estate of a postage stamp and dissipate next to nothing, as in
<500mW. If it doesn't it slowly cooks out the relay box, as has happened
on the heat pump controller we had in Europe."

... and "No supply rail". Of course 24V industrial means it should take
40V peaks.
Inductance?

"...held there for a msec or so"?? How many "so's"??

Not determined yet, could be 3, 4, 5msec. In a circuit like this it
should not matter, I just make that time selectable with a component
such as R or C. It is something that will most likely have to be changed
later. Usually the allowed tolerance is large. If it had to be under 10%
there would always be the old CD4060 which nowadays even comes in TSSOP.

Inductance??

Around 10mH or a bit higher, also not cast in concrete yet.

With 24 volts that won't get to 3A "in a fraction of a msec".

Now do you see why I'm picky about conciseness ?

How do you rationalize 10mH with your speed "specs"?

As vague as your description has been, you would have been better off
to specify _FUNCTION_ desired.

Vague, vague, vague will get you nothing but criticism ;-)

However, in
the end this doesn't matter much, it's just the principle of operation.
No problem to do it in all discretes but the layouter will have a cow.
Why camn't anyone make a L295 in CMOS? With all that automotive stuff
nowadays there should be a market for a 60V rated device. Maybe one of
your clients could make a profit there, although since Lopez they may
not like automotive much anymore.

I haven't kept up with the automotive market, but the trend is to go
to 24V automotive systems. So I would expect some high side drivers
available for "24V" (nominal battery... ~31V maximum plus surges).

Just fixed our wireless doorbell extender. Major "de-crudding", a D cell
had decided to take a significant pee. Why is it that out of a set of
leak-proof alkalines one has to leak?

It will always be a cell which is against the battery holder spring,
so it can ruin the spring ;-)

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine Sometimes I even put it in the food

 

[Joerg]

Jim Thompson wrote:
[...]

"It's not cast in concrete yet but something like 24V industry rail with
tons of noise on there, coil about an ohm so it can ramp up to 3A or so
in a fraction of a msec, then must be held there for a msec or so, then
ramped down to hold (roughly 1/5th) in 100usec. All this must fit onto
the real estate of a postage stamp and dissipate next to nothing, as in
<500mW. If it doesn't it slowly cooks out the relay box, as has happened
on the heat pump controller we had in Europe."

... and "No supply rail". Of course 24V industrial means it should take
40V peaks.
Inductance?

"...held there for a msec or so"?? How many "so's"??

Not determined yet, could be 3, 4, 5msec. In a circuit like this it
should not matter, I just make that time selectable with a component
such as R or C. It is something that will most likely have to be changed
later. Usually the allowed tolerance is large. If it had to be under 10%
there would always be the old CD4060 which nowadays even comes in TSSOP.
Inductance??

Around 10mH or a bit higher, also not cast in concrete yet.

With 24 volts that won't get to 3A "in a fraction of a msec".

Now do you see why I'm picky about conciseness ?

How do you rationalize 10mH with your speed "specs"?

Hint: There are ways to temporarily make more volts with another
inductor, a FET, an oscillator ... ;-)

But I probably won't have to.

As vague as your description has been, you would have been better off
to specify _FUNCTION_ desired.

Vague, vague, vague will get you nothing but criticism ;-)

I never criticize my clients for that because they usually have good
reasons, meaning stuff hasn't been designed yet on their side. Maybe
that's different in the IC designer world but in my world that is very
normal. And yeah, it has happened that I got a call late in the game and
someone said "S..t, I think we actually do need to run the Doppler
simultaneously with B-mode" (that meant a scrapped vacation trip and a
whole new power supply design).

I haven't kept up with the automotive market, but the trend is to go
to 24V automotive systems. So I would expect some high side drivers
available for "24V" (nominal battery... ~31V maximum plus surges).

That would require 100V devices to handle load dumps and whatnot. On 12V
systems it's customary to shoot for 50-60V. I mean at good manufacturers.

It will always be a cell which is against the battery holder spring,
so it can ruin the spring ;-)

Of course. It was the one where the spring is at the bottom and that's
the side it leaked out. Anyway, fixed, and it ringeth

Other question, of course rhetorical as well: When do designers learn to
place batteries sideways so leak damage is minimized?

What blew me away was when I bought this Dimango bell 10 years ago. $20,
with (!) transmitter that goes into the main doorbell. It's a pretty
elaborate receiver, lots of through-hole parts on phenolic. A
masterpiece of cost efficient design.

 

[Son of a Sea Cook]

Need to drive a >30V (several amps) coil but back off once it's on to
avoid a cook-out. This needs to be PWM'd because the drive electronics
can't get hot at all. So I looked at coil driver chips. DRV103 and the
like. Many can't stomach more than 32V which is borderline for me and
the saturation voltage respectively Rdson values, well, yawn. Nothing to
write home about.

Did I miss a really good chip? Else I'll just roll my own.

We used a simple audio class D amp chip to drive one of our designs
once. A simple mono amp from Fry's comes to mind.

Otherwise FETs or IGBTs

 

[Jim Thompson]

Jim Thompson wrote:
Jim Thompson wrote:
[...]

"It's not cast in concrete yet but something like 24V industry rail with
tons of noise on there, coil about an ohm so it can ramp up to 3A or so
in a fraction of a msec, then must be held there for a msec or so, then
ramped down to hold (roughly 1/5th) in 100usec. All this must fit onto
the real estate of a postage stamp and dissipate next to nothing, as in
<500mW. If it doesn't it slowly cooks out the relay box, as has happened
on the heat pump controller we had in Europe."

... and "No supply rail". Of course 24V industrial means it should take
40V peaks.
Inductance?

"...held there for a msec or so"?? How many "so's"??

Not determined yet, could be 3, 4, 5msec. In a circuit like this it
should not matter, I just make that time selectable with a component
such as R or C. It is something that will most likely have to be changed
later. Usually the allowed tolerance is large. If it had to be under 10%
there would always be the old CD4060 which nowadays even comes in TSSOP.
Inductance??

Around 10mH or a bit higher, also not cast in concrete yet.

With 24 volts that won't get to 3A "in a fraction of a msec".

Now do you see why I'm picky about conciseness ?

How do you rationalize 10mH with your speed "specs"?

Hint: There are ways to temporarily make more volts with another
inductor, a FET, an oscillator ... ;-)

But I probably won't have to.

But you implied that 24V was all you had available, and it only popped
up when you want to activate the "thing", thus the "pump" response
time would shoot you in the foot.

I never criticize my clients for that because they usually have good
reasons, meaning stuff hasn't been designed yet on their side. Maybe
that's different in the IC designer world but in my world that is very
normal. And yeah, it has happened that I got a call late in the game and
someone said "S..t, I think we actually do need to run the Doppler
simultaneously with B-mode" (that meant a scrapped vacation trip and a
whole new power supply design).

I always dig into the client's system concept and recommend
improvements in their architecture... often with significant cost and
area savings... AND performance improvements ;-)

[snip]

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine Sometimes I even put it in the food

 

[Joerg]

Jim Thompson wrote:
Jim Thompson wrote:
[...]

"It's not cast in concrete yet but something like 24V industry rail with
tons of noise on there, coil about an ohm so it can ramp up to 3A or so
in a fraction of a msec, then must be held there for a msec or so, then
ramped down to hold (roughly 1/5th) in 100usec. All this must fit onto
the real estate of a postage stamp and dissipate next to nothing, as in
<500mW. If it doesn't it slowly cooks out the relay box, as has happened
on the heat pump controller we had in Europe."

... and "No supply rail". Of course 24V industrial means it should take
40V peaks.
Inductance?

"...held there for a msec or so"?? How many "so's"??

Not determined yet, could be 3, 4, 5msec. In a circuit like this it
should not matter, I just make that time selectable with a component
such as R or C. It is something that will most likely have to be changed
later. Usually the allowed tolerance is large. If it had to be under 10%
there would always be the old CD4060 which nowadays even comes in TSSOP.
Inductance??

Around 10mH or a bit higher, also not cast in concrete yet.
With 24 volts that won't get to 3A "in a fraction of a msec".

Now do you see why I'm picky about conciseness ?

How do you rationalize 10mH with your speed "specs"?

Hint: There are ways to temporarily make more volts with another
inductor, a FET, an oscillator ... ;-)

But I probably won't have to.

But you implied that 24V was all you had available, and it only popped
up when you want to activate the "thing", thus the "pump" response
time would shoot you in the foot.

I can make a pump start muy pronto

<brag_mode>
A while ago a 6V converter had to be ready to supply full bore in under
200usec after applying the DC input voltage. It was ready to salute in
120usec. That evoked a "shazam!" from the guys.

I always dig into the client's system concept and recommend
improvements in their architecture... often with significant cost and
area savings... AND performance improvements ;-)

Sure, so do I. I had brought up exactly this question of whether
simultaneous operation was needed when the project started. The answer
was that it would not be required. "Are you sure?" ... "Yes." ...
"Hundred percent?" ... "YES!" ... Then, however, the marketeers found
out that the competition ...

Sometimes this goes past EE stuff. Repeated inquiries at to mechanical
vibrations were answered with "not a concern", to the point where one
guy said in a somewhat irritated tone "Now let's not worry about this
anymore!". Guess which issue caused a significant re-design effort a few
months later?

 

[Jim Thompson]

Jim Thompson wrote: [snip]

I always dig into the client's system concept and recommend
improvements in their architecture... often with significant cost and
area savings... AND performance improvements ;-)

Sure, so do I. I had brought up exactly this question of whether
simultaneous operation was needed when the project started. The answer
was that it would not be required. "Are you sure?" ... "Yes." ...
"Hundred percent?" ... "YES!" ... Then, however, the marketeers found
out that the competition ...

Sometimes this goes past EE stuff. Repeated inquiries at to mechanical
vibrations were answered with "not a concern", to the point where one
guy said in a somewhat irritated tone "Now let's not worry about this
anymore!". Guess which issue caused a significant re-design effort a few
months later?

I have one of those. Every few months, "Can we change (or add) this?"

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine Sometimes I even put it in the food

 

[Joerg]

Jim Thompson wrote: [snip]

I always dig into the client's system concept and recommend
improvements in their architecture... often with significant cost and
area savings... AND performance improvements ;-)

Sure, so do I. I had brought up exactly this question of whether
simultaneous operation was needed when the project started. The answer
was that it would not be required. "Are you sure?" ... "Yes." ...
"Hundred percent?" ... "YES!" ... Then, however, the marketeers found
out that the competition ...

Sometimes this goes past EE stuff. Repeated inquiries at to mechanical
vibrations were answered with "not a concern", to the point where one
guy said in a somewhat irritated tone "Now let's not worry about this
anymore!". Guess which issue caused a significant re-design effort a few
months later?

I have one of those. Every few months, "Can we change (or add) this?"

Since I assume you are paid time and materials change orders should be
fine. Think how many bottles of Syrah that brings in

Now it's off onto the (hot) roof, trying to find a compromise with the
antenna between DTV multipath, intermod choking of all this consumer
stuff, dang, wish my luck.

The worst is our local station, >50dBuV on the analyzer all the time but
often "no signal" on the TV at night. Pathetic.

 

[Jim Thompson]

Jim Thompson wrote: [snip]
I always dig into the client's system concept and recommend
improvements in their architecture... often with significant cost and
area savings... AND performance improvements ;-)

Sure, so do I. I had brought up exactly this question of whether
simultaneous operation was needed when the project started. The answer
was that it would not be required. "Are you sure?" ... "Yes." ...
"Hundred percent?" ... "YES!" ... Then, however, the marketeers found
out that the competition ...

Sometimes this goes past EE stuff. Repeated inquiries at to mechanical
vibrations were answered with "not a concern", to the point where one
guy said in a somewhat irritated tone "Now let's not worry about this
anymore!". Guess which issue caused a significant re-design effort a few
months later?

I have one of those. Every few months, "Can we change (or add) this?"

Since I assume you are paid time and materials change orders should be
fine. Think how many bottles of Syrah that brings in

Now it's off onto the (hot) roof, trying to find a compromise with the
antenna between DTV multipath, intermod choking of all this consumer
stuff, dang, wish my luck.

The worst is our local station, >50dBuV on the analyzer all the time but
often "no signal" on the TV at night. Pathetic.

Sometimes "hot" is good. Went out this morning to spray rabbit
repellent, to be greeted by sage full of bees (the sage is in bloom).

I know that bees don't really like extremes of heat, so I waited until
noon... not a bee in sight ;-)

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine Sometimes I even put it in the food

 

[Joerg]

[...]

Now it's off onto the (hot) roof, trying to find a compromise with the
antenna between DTV multipath, intermod choking of all this consumer
stuff, dang, wish my luck.

The worst is our local station, >50dBuV on the analyzer all the time but
often "no signal" on the TV at night. Pathetic.

Sometimes "hot" is good. Went out this morning to spray rabbit
repellent, to be greeted by sage full of bees (the sage is in bloom).

I know that bees don't really like extremes of heat, so I waited until
noon... not a bee in sight ;-)

Be careful, my wife got stung by a bee earlier this week.

I did burn my right hand up there a bit. Decra metal roof with embedded
granules, looks like shake but touch it and ... phsssss. Anyhow no DTV
improvement to be had. I can boost here and there, looks beautiful on
the analyzer plot but then the TVs fall off the rocker. Maybe the
engineers didn't know what an IP3 is :-(

Now, off into the pool. Yeehaw!

 

[Ross Herbert]

:Need to drive a >30V (several amps) coil but back off once it's on to
:avoid a cook-out. This needs to be PWM'd because the drive electronics
:can't get hot at all. So I looked at coil driver chips. DRV103 and the
:like. Many can't stomach more than 32V which is borderline for me and
:the saturation voltage respectively Rdson values, well, yawn. Nothing to
:write home about.
:
id I miss a really good chip? Else I'll just roll my own.


Have you considered the Ericsson range of drivers? eg. PBD354x/1
https://www1.elfa.se/data1/wwwroot/webroot/Z_DATA/07335482.pdf (source)
http://datasheet.digchip.com/000/000-0-PBD3545-1.pdf (sink)

 

[Nico Coesel]

Jim Thompson a écrit :

Jim Thompson wrote:
[...]

"It's not cast in concrete yet but something like 24V industry rail with
tons of noise on there, coil about an ohm so it can ramp up to 3A or so
in a fraction of a msec, then must be held there for a msec or so, then
ramped down to hold (roughly 1/5th) in 100usec. All this must fit onto
the real estate of a postage stamp and dissipate next to nothing, as in
<500mW. If it doesn't it slowly cooks out the relay box, as has happened
on the heat pump controller we had in Europe."

... and "No supply rail". Of course 24V industrial means it should take
40V peaks.
Inductance?

"...held there for a msec or so"?? How many "so's"??

Not determined yet, could be 3, 4, 5msec. In a circuit like this it
should not matter, I just make that time selectable with a component
such as R or C. It is something that will most likely have to be changed
later. Usually the allowed tolerance is large. If it had to be under 10%
there would always be the old CD4060 which nowadays even comes in TSSOP.

Inductance??

...Jim Thompson

voltage*rise_time/current

Of course. I'm just trying to get Joerg to be concise. As usual he's
as vague and side-stepping as Larkin ;-)

I did a similar design for a valve (open it and then ramp down the
current to keep it from overheating). This stuff doesn't specify
inductance. Heck, they could connect anything to it. IMHO what Joerg
wants is something that uses nearly no power, does the ramp-thing and
is idiot proof (short circuit, reverse polarity). Industrial relay
stuff is never concise.

I think the specs are bogus given the surrounding hardware but I know
that some customers have no idea about what they spec and telling them
they are wrong scares them away.

 

[Fred Bartoli]

Nico Coesel a écrit :

Jim Thompson a écrit :
Jim Thompson wrote:
[...]

"It's not cast in concrete yet but something like 24V industry rail with
tons of noise on there, coil about an ohm so it can ramp up to 3A or so
in a fraction of a msec, then must be held there for a msec or so, then
ramped down to hold (roughly 1/5th) in 100usec. All this must fit onto
the real estate of a postage stamp and dissipate next to nothing, as in
<500mW. If it doesn't it slowly cooks out the relay box, as has happened
on the heat pump controller we had in Europe."

... and "No supply rail". Of course 24V industrial means it should take
40V peaks.
Inductance?

"...held there for a msec or so"?? How many "so's"??

Not determined yet, could be 3, 4, 5msec. In a circuit like this it
should not matter, I just make that time selectable with a component
such as R or C. It is something that will most likely have to be changed
later. Usually the allowed tolerance is large. If it had to be under 10%
there would always be the old CD4060 which nowadays even comes in TSSOP.
Inductance??

...Jim Thompson
voltage*rise_time/current

Of course. I'm just trying to get Joerg to be concise. As usual he's
as vague and side-stepping as Larkin ;-)

I did a similar design for a valve (open it and then ramp down the
current to keep it from overheating). This stuff doesn't specify
inductance. Heck, they could connect anything to it. IMHO what Joerg
wants is something that uses nearly no power, does the ramp-thing and
is idiot proof (short circuit, reverse polarity). Industrial relay
stuff is never concise.

I think the specs are bogus given the surrounding hardware but I know
that some customers have no idea about what they spec and telling them
they are wrong scares them away.

Then don't tell them they're wrong.

I such case, I tell them I can help them have a better understanding of
how this works. And that that have a cost too

 

[Jim Thompson]

Nico Coesel a écrit :

[snip]

I did a similar design for a valve (open it and then ramp down the
current to keep it from overheating). This stuff doesn't specify
inductance. Heck, they could connect anything to it. IMHO what Joerg
wants is something that uses nearly no power, does the ramp-thing and
is idiot proof (short circuit, reverse polarity). Industrial relay
stuff is never concise.

I think the specs are bogus given the surrounding hardware but I know
that some customers have no idea about what they spec and telling them
they are wrong scares them away.

Then don't tell them they're wrong.

I such case, I tell them I can help them have a better understanding of
how this works. And that that have a cost too

I often resort to asking the client, "Measure this, measure that,
under these conditions..."

I only tell the client they are stupid if they get belligerent. If
they persist, I tell them to get someone else. Amazing how many come
back... often _years_ and _millions_of_dollars_ later, to have me redo
their chip design ;-)

Admittedly, Joerg is often dealing with medical types. When I wrote
up an instruction manual for blood flow apparatus I built as my
Bachelor's thesis (joint MIT and Harvard School of Medicine), I had my
wife proof read it. Her comment, "It's seems at an awfully juvenile
level". I told her it was necessary when dealing with doctors ;-)

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

"The American Republic will endure until the day Congress
discovers that it can bribe the public with the public's money."

- Alexis de Tocqueville

 

[Jim Thompson]

:Need to drive a >30V (several amps) coil but back off once it's on to
:avoid a cook-out. This needs to be PWM'd because the drive electronics
:can't get hot at all. So I looked at coil driver chips. DRV103 and the
:like. Many can't stomach more than 32V which is borderline for me and
:the saturation voltage respectively Rdson values, well, yawn. Nothing to
:write home about.
:
id I miss a really good chip? Else I'll just roll my own.


Have you considered the Ericsson range of drivers? eg. PBD354x/1
https://www1.elfa.se/data1/wwwroot/webroot/Z_DATA/07335482.pdf (source)
http://datasheet.digchip.com/000/000-0-PBD3545-1.pdf (sink)

Those look good, though Joerg implies he has no other power than the
"24V" when it's applied.

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

All that is necessary for the triumph of evil
is that good men do nothing.
-Edmund Burke

 

[Joerg]

Joerg, I suggest you look at IR's IRS2153 or 21531 self-oscillating
high-voltage half-bridge ICs. These generate their own low-voltage
supply from the coil supply, and they have a 555-like oscillator that
you can easily trick into continuous on or off states, besides PWM.

Thanks, Win. Very interesting chip. However, for an initial continuous
on state of the upper FET the boost cap won't come up because for some
reason they just hung that onto the drive for the lower FET. But there
may be options with connecting the coil to V+. Makes current sense on it
a bear though.

Do you know which market they normally serve? The data sheet is a bit
mum about that. What's really nice is that the chip goes to 100kHz.
Should be high enough to muffle conducted noise at least to some extent.
If I don't they'd read me the riot act ;-)

 

[Joerg]

Thanks, Ross. Vcesat looks highish though. The PLCC pack is the only one
I could fit and it would probably unsolder itself ;-)

Well, not literally since it has a shutdown funtion but that would be
quite disruptive as well.

Those look good, though Joerg implies he has no other power than the
"24V" when it's applied.

Yep, only that switched line plus the usual spikes, surges and such on
it. If I had enough real estate I'd be almost home. Same if I had
unfettered access to a nice HV foundry, plus a wad of cash to pay for
masks and NRE

 

[Joerg]

I'm assuming you'd drive the coil to ground. That way you start out
with the output low and with the coil off, and at that time the hi-
side
flying cap gets charged, ready for your first turn-on.

Actually I must ramp the coil current to the hilt the instant the rail
comes up. In cases like this the rail is the switched line and also the
only power source. Wanting an extra DC power line to come in is almost
like renting a backhoe and starting to trench

A fixed duty-cycle PWM half-bridge setup can be seen as an
unregulated
DC transformer, and there are potentially many types of applications
for
that. IR actually has an entire series of similar parts. Their
press release
claimed the 2153 was meant for flourescent bulb ballasts. But it
seems
inadequate to the task to me. Their newer ICs, such as the IRS2530,
do
more of the job, which is fairly complicated.

I have to go through their stuff one more time, maybe they really have
something useful. Usually Supertex comes to the rescue in such cases but
not this time. They have only the HV9901 and it wouldn't buy me that
much in space savings versus discretes. OTOH, cost is not the driving
factor on this design, within reason, meaning no chip design. Ain't that
nice for a change?

Sometimes I envy guys like Jim but I had that luxury only twice: You
shop around for HV processes and then piece it all together on your
computation machine. After I did a design on a 60V process I was so spoiled.