On Wed, 23 Oct 2013 17:03:08 -0700, Jim Thompson
<To-Email-Use-The-Envelope-Icon@On-My-Web-Site.com> wrote:
>On Wed, 23 Oct 2013 19:26:01 -0400, krw@attt.bizz wrote:
>
>>On Wed, 23 Oct 2013 16:49:56 -0400, Phil Hobbs
>><pcdhSpamMeSenseless@electrooptical.net> wrote:
>>
>>>On 10/23/2013 04:27 PM, miso wrote:
>>>> On 10/20/2013 12:24 PM, John Larkin wrote:
>>>>> On Sun, 20 Oct 2013 14:07:41 -0400, Phil Hobbs
>>>>> <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>>>
>>>>>> On 10/20/2013 2:03 PM, John Larkin wrote:
>>>>>>>
>>>>>>>
>>>>>>> Lots of CMOS gates and such are tolerant of pin voltages above Vcc,
>>>>>>> typically
>>>>>>> rated for 7 volts max. Does anybody know what sort of input circuits
>>>>>>> are
>>>>>>> typically used? Does it behave like a zener? To ground?
>>>>>>>
>>>>>>> Some parts specify an allowed input curent in the pullup direction,
>>>>>>> some don't.
>>>>>>>
>>>>>>> I'm designing a really tiny board, and an opamp can potentially
>>>>>>> overdrive an
>>>>>>> analog mux, an FSA3157. I can current limit it with a "free"
>>>>>>> resistor, in a
>>>>>>> pack, but a diode would take more room.
>>>>>>
>>>>>> Overdriving an analogue mux can have a lot of nasty effects without
>>>>>> blowing up the part, e.g. connecting all the inputs together. How about
>>>>>> a nice voltage divider to prevent the overdrive in the first place?
>>>>>>
>>>>>
>>>>> I don't mind the mux getting weird, I just don't want to fry it. If I
>>>>> absolutely
>>>>> have to add a part, it could be a tiny schottky diode to +5. I just
>>>>> begrudge
>>>>> every part on this schematic; the PCB will be under 2 square inches.
>>>>> I'll be
>>>>> using a dual opamp and the other half needs to be powered from +12, so
>>>>> the one
>>>>> that I'm gain-switching x1/x10 with the mux is +12 too.
>>>>>
>>>>> But I've been generally interested in finding out what's inside those
>>>>> "5 volt
>>>>> tolerant" cmos parts. The positive direction must have some sort of
>>>>> zener-y
>>>>> thing, for ESD.
>>>>>
>>>>>
>>>>
>>>> ESD in the negative direction is with a body diode. The positive
>>>> direction depends on the device breaking down, i.e. snap back, and of
>>>> course surviving. This is very old technology dating back to the early
>>>> 90s, but not really advertised much until mixed voltages became more
>>>> common. You may recall I mentioned this in your "Don't buy Maxim thread"
>>>> since the customer fucks up a lot with multiple power supplies, so
>>>> leaving the positive rail diode off the chip is one less thing for the
>>>> customer to screw up.
>>>>
>>>> Personally, I don't think the snap back Nfet is as good as the diodes to
>>>> both rails, but it is a matter of customer confidence. Remember, 99.9%
>>>> of the time, the customer fucks up, but they believe that the
>>>> manufacturer is the problem. [The manufacturer is selling billions of
>>>> dollars of compoenents, so obviously some customers know what they are
>>>> doing.] If the customer is latching up parts due to a sequencing issue
>>>> with multiple voltages on board, they will design out your part (even
>>>> though they fucked up), they will bad mouth your company (even though
>>>> they fucked up), or do a combination of both. Customer shit is like ice
>>>> cream, while the manufacturer's shit is ..well shit!
>>>>
>>>> Thus all sorts of protection go into chips to keep the moronic customer
>>>> at bay. You have to assume some oaf will short adjacent pins together
>>>> with a scope probe, so you think about what will happen under those
>>>> conditions. If you have a power device on chip, you put in some short
>>>> circuit protection just in case the customer shorts it to a rail, since
>>>> their clumsiness is of course the manufacturer's fault.
>>>>
>>>> You can't believe the amount of work that goes into a power on reset /
>>>> undervoltage lockout scheme. Some customers are so fucking cheap they
>>>> design their owner DC/DC rather than use a good controller chip, so you
>>>> need to be ready for really bad power supply start up issues. Never ever
>>>> assume power supply power up will be monotonic.
>>>>
>>>>
>>>
>>>Well, of course we poor benighted customers never realized that our
>>>problems were so much less important than yours. I apologize deeply and
>>>promise that I will never, never complain again, even when some poor
>>>victimized IC house fails to ship the promised wonder-chip for the Nth
>>>time in a row.
>>>
>>>Could I adjust your pillow to make you more comfortable? How about a
>>>nice cup of tea?
>>
>>Miso worked for Maxim so understand where his head is.
>
>Same place as yours ?>:-}
I thought you killfiled me. Please do.
Reply by Phil Hobbs●October 23, 20132013-10-23
On 10/23/2013 7:26 PM, krw@attt.bizz wrote:
> On Wed, 23 Oct 2013 16:49:56 -0400, Phil Hobbs
> <pcdhSpamMeSenseless@electrooptical.net> wrote:
>
>> On 10/23/2013 04:27 PM, miso wrote:
>>> On 10/20/2013 12:24 PM, John Larkin wrote:
>>>> On Sun, 20 Oct 2013 14:07:41 -0400, Phil Hobbs
>>>> <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>>
>>>>> On 10/20/2013 2:03 PM, John Larkin wrote:
>>>>>>
>>>>>>
>>>>>> Lots of CMOS gates and such are tolerant of pin voltages above Vcc,
>>>>>> typically
>>>>>> rated for 7 volts max. Does anybody know what sort of input circuits
>>>>>> are
>>>>>> typically used? Does it behave like a zener? To ground?
>>>>>>
>>>>>> Some parts specify an allowed input curent in the pullup direction,
>>>>>> some don't.
>>>>>>
>>>>>> I'm designing a really tiny board, and an opamp can potentially
>>>>>> overdrive an
>>>>>> analog mux, an FSA3157. I can current limit it with a "free"
>>>>>> resistor, in a
>>>>>> pack, but a diode would take more room.
>>>>>
>>>>> Overdriving an analogue mux can have a lot of nasty effects without
>>>>> blowing up the part, e.g. connecting all the inputs together. How about
>>>>> a nice voltage divider to prevent the overdrive in the first place?
>>>>>
>>>>
>>>> I don't mind the mux getting weird, I just don't want to fry it. If I
>>>> absolutely
>>>> have to add a part, it could be a tiny schottky diode to +5. I just
>>>> begrudge
>>>> every part on this schematic; the PCB will be under 2 square inches.
>>>> I'll be
>>>> using a dual opamp and the other half needs to be powered from +12, so
>>>> the one
>>>> that I'm gain-switching x1/x10 with the mux is +12 too.
>>>>
>>>> But I've been generally interested in finding out what's inside those
>>>> "5 volt
>>>> tolerant" cmos parts. The positive direction must have some sort of
>>>> zener-y
>>>> thing, for ESD.
>>>>
>>>>
>>>
>>> ESD in the negative direction is with a body diode. The positive
>>> direction depends on the device breaking down, i.e. snap back, and of
>>> course surviving. This is very old technology dating back to the early
>>> 90s, but not really advertised much until mixed voltages became more
>>> common. You may recall I mentioned this in your "Don't buy Maxim thread"
>>> since the customer fucks up a lot with multiple power supplies, so
>>> leaving the positive rail diode off the chip is one less thing for the
>>> customer to screw up.
>>>
>>> Personally, I don't think the snap back Nfet is as good as the diodes to
>>> both rails, but it is a matter of customer confidence. Remember, 99.9%
>>> of the time, the customer fucks up, but they believe that the
>>> manufacturer is the problem. [The manufacturer is selling billions of
>>> dollars of compoenents, so obviously some customers know what they are
>>> doing.] If the customer is latching up parts due to a sequencing issue
>>> with multiple voltages on board, they will design out your part (even
>>> though they fucked up), they will bad mouth your company (even though
>>> they fucked up), or do a combination of both. Customer shit is like ice
>>> cream, while the manufacturer's shit is ..well shit!
>>>
>>> Thus all sorts of protection go into chips to keep the moronic customer
>>> at bay. You have to assume some oaf will short adjacent pins together
>>> with a scope probe, so you think about what will happen under those
>>> conditions. If you have a power device on chip, you put in some short
>>> circuit protection just in case the customer shorts it to a rail, since
>>> their clumsiness is of course the manufacturer's fault.
>>>
>>> You can't believe the amount of work that goes into a power on reset /
>>> undervoltage lockout scheme. Some customers are so fucking cheap they
>>> design their owner DC/DC rather than use a good controller chip, so you
>>> need to be ready for really bad power supply start up issues. Never ever
>>> assume power supply power up will be monotonic.
>>>
>>>
>>
>> Well, of course we poor benighted customers never realized that our
>> problems were so much less important than yours. I apologize deeply and
>> promise that I will never, never complain again, even when some poor
>> victimized IC house fails to ship the promised wonder-chip for the Nth
>> time in a row.
>>
>> Could I adjust your pillow to make you more comfortable? How about a
>> nice cup of tea?
>
> Miso worked for Maxim so understand where his head is.
>
Gentle mockery is harder to do on Usenet, because everybody's wearing
their Nomex shorts. ;)
Cheers
Phil Hobbs
--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics
160 North State Road #203
Briarcliff Manor NY 10510 USA
+1 845 480 2058
hobbs at electrooptical dot net
http://electrooptical.net
Reply by Jim Thompson●October 23, 20132013-10-23
On Wed, 23 Oct 2013 17:01:15 -0700, John Larkin
<jlarkin@highlandtechnology.com> wrote:
>On Wed, 23 Oct 2013 19:26:01 -0400, krw@attt.bizz wrote:
>
>>On Wed, 23 Oct 2013 16:49:56 -0400, Phil Hobbs
>><pcdhSpamMeSenseless@electrooptical.net> wrote:
>
>
>>>Well, of course we poor benighted customers never realized that our
>>>problems were so much less important than yours. I apologize deeply and
>>>promise that I will never, never complain again, even when some poor
>>>victimized IC house fails to ship the promised wonder-chip for the Nth
>>>time in a row.
>>>
>>>Could I adjust your pillow to make you more comfortable? How about a
>>>nice cup of tea?
>>
>>Miso worked for Maxim so understand where his head is.
>
>
>Never Buy Miso!
Never buy Larkin?
...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 Skype: Contacts Only | |
| Voice:(480)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.
Reply by Jim Thompson●October 23, 20132013-10-23
On Wed, 23 Oct 2013 19:26:01 -0400, krw@attt.bizz wrote:
>On Wed, 23 Oct 2013 16:49:56 -0400, Phil Hobbs
><pcdhSpamMeSenseless@electrooptical.net> wrote:
>
>>On 10/23/2013 04:27 PM, miso wrote:
>>> On 10/20/2013 12:24 PM, John Larkin wrote:
>>>> On Sun, 20 Oct 2013 14:07:41 -0400, Phil Hobbs
>>>> <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>>
>>>>> On 10/20/2013 2:03 PM, John Larkin wrote:
>>>>>>
>>>>>>
>>>>>> Lots of CMOS gates and such are tolerant of pin voltages above Vcc,
>>>>>> typically
>>>>>> rated for 7 volts max. Does anybody know what sort of input circuits
>>>>>> are
>>>>>> typically used? Does it behave like a zener? To ground?
>>>>>>
>>>>>> Some parts specify an allowed input curent in the pullup direction,
>>>>>> some don't.
>>>>>>
>>>>>> I'm designing a really tiny board, and an opamp can potentially
>>>>>> overdrive an
>>>>>> analog mux, an FSA3157. I can current limit it with a "free"
>>>>>> resistor, in a
>>>>>> pack, but a diode would take more room.
>>>>>
>>>>> Overdriving an analogue mux can have a lot of nasty effects without
>>>>> blowing up the part, e.g. connecting all the inputs together. How about
>>>>> a nice voltage divider to prevent the overdrive in the first place?
>>>>>
>>>>
>>>> I don't mind the mux getting weird, I just don't want to fry it. If I
>>>> absolutely
>>>> have to add a part, it could be a tiny schottky diode to +5. I just
>>>> begrudge
>>>> every part on this schematic; the PCB will be under 2 square inches.
>>>> I'll be
>>>> using a dual opamp and the other half needs to be powered from +12, so
>>>> the one
>>>> that I'm gain-switching x1/x10 with the mux is +12 too.
>>>>
>>>> But I've been generally interested in finding out what's inside those
>>>> "5 volt
>>>> tolerant" cmos parts. The positive direction must have some sort of
>>>> zener-y
>>>> thing, for ESD.
>>>>
>>>>
>>>
>>> ESD in the negative direction is with a body diode. The positive
>>> direction depends on the device breaking down, i.e. snap back, and of
>>> course surviving. This is very old technology dating back to the early
>>> 90s, but not really advertised much until mixed voltages became more
>>> common. You may recall I mentioned this in your "Don't buy Maxim thread"
>>> since the customer fucks up a lot with multiple power supplies, so
>>> leaving the positive rail diode off the chip is one less thing for the
>>> customer to screw up.
>>>
>>> Personally, I don't think the snap back Nfet is as good as the diodes to
>>> both rails, but it is a matter of customer confidence. Remember, 99.9%
>>> of the time, the customer fucks up, but they believe that the
>>> manufacturer is the problem. [The manufacturer is selling billions of
>>> dollars of compoenents, so obviously some customers know what they are
>>> doing.] If the customer is latching up parts due to a sequencing issue
>>> with multiple voltages on board, they will design out your part (even
>>> though they fucked up), they will bad mouth your company (even though
>>> they fucked up), or do a combination of both. Customer shit is like ice
>>> cream, while the manufacturer's shit is ..well shit!
>>>
>>> Thus all sorts of protection go into chips to keep the moronic customer
>>> at bay. You have to assume some oaf will short adjacent pins together
>>> with a scope probe, so you think about what will happen under those
>>> conditions. If you have a power device on chip, you put in some short
>>> circuit protection just in case the customer shorts it to a rail, since
>>> their clumsiness is of course the manufacturer's fault.
>>>
>>> You can't believe the amount of work that goes into a power on reset /
>>> undervoltage lockout scheme. Some customers are so fucking cheap they
>>> design their owner DC/DC rather than use a good controller chip, so you
>>> need to be ready for really bad power supply start up issues. Never ever
>>> assume power supply power up will be monotonic.
>>>
>>>
>>
>>Well, of course we poor benighted customers never realized that our
>>problems were so much less important than yours. I apologize deeply and
>>promise that I will never, never complain again, even when some poor
>>victimized IC house fails to ship the promised wonder-chip for the Nth
>>time in a row.
>>
>>Could I adjust your pillow to make you more comfortable? How about a
>>nice cup of tea?
>
>Miso worked for Maxim so understand where his head is.
Same place as yours ?>:-}
...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 Skype: Contacts Only | |
| Voice:(480)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.
Reply by John Larkin●October 23, 20132013-10-23
On Wed, 23 Oct 2013 19:26:01 -0400, krw@attt.bizz wrote:
>On Wed, 23 Oct 2013 16:49:56 -0400, Phil Hobbs
><pcdhSpamMeSenseless@electrooptical.net> wrote:
>>Well, of course we poor benighted customers never realized that our
>>problems were so much less important than yours. I apologize deeply and
>>promise that I will never, never complain again, even when some poor
>>victimized IC house fails to ship the promised wonder-chip for the Nth
>>time in a row.
>>
>>Could I adjust your pillow to make you more comfortable? How about a
>>nice cup of tea?
>
>Miso worked for Maxim so understand where his head is.
Never Buy Miso!
--
John Larkin Highland Technology, Inc
jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro acquisition and simulation
Reply by ●October 23, 20132013-10-23
On Wed, 23 Oct 2013 16:49:56 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:
>On 10/23/2013 04:27 PM, miso wrote:
>> On 10/20/2013 12:24 PM, John Larkin wrote:
>>> On Sun, 20 Oct 2013 14:07:41 -0400, Phil Hobbs
>>> <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>
>>>> On 10/20/2013 2:03 PM, John Larkin wrote:
>>>>>
>>>>>
>>>>> Lots of CMOS gates and such are tolerant of pin voltages above Vcc,
>>>>> typically
>>>>> rated for 7 volts max. Does anybody know what sort of input circuits
>>>>> are
>>>>> typically used? Does it behave like a zener? To ground?
>>>>>
>>>>> Some parts specify an allowed input curent in the pullup direction,
>>>>> some don't.
>>>>>
>>>>> I'm designing a really tiny board, and an opamp can potentially
>>>>> overdrive an
>>>>> analog mux, an FSA3157. I can current limit it with a "free"
>>>>> resistor, in a
>>>>> pack, but a diode would take more room.
>>>>
>>>> Overdriving an analogue mux can have a lot of nasty effects without
>>>> blowing up the part, e.g. connecting all the inputs together. How about
>>>> a nice voltage divider to prevent the overdrive in the first place?
>>>>
>>>
>>> I don't mind the mux getting weird, I just don't want to fry it. If I
>>> absolutely
>>> have to add a part, it could be a tiny schottky diode to +5. I just
>>> begrudge
>>> every part on this schematic; the PCB will be under 2 square inches.
>>> I'll be
>>> using a dual opamp and the other half needs to be powered from +12, so
>>> the one
>>> that I'm gain-switching x1/x10 with the mux is +12 too.
>>>
>>> But I've been generally interested in finding out what's inside those
>>> "5 volt
>>> tolerant" cmos parts. The positive direction must have some sort of
>>> zener-y
>>> thing, for ESD.
>>>
>>>
>>
>> ESD in the negative direction is with a body diode. The positive
>> direction depends on the device breaking down, i.e. snap back, and of
>> course surviving. This is very old technology dating back to the early
>> 90s, but not really advertised much until mixed voltages became more
>> common. You may recall I mentioned this in your "Don't buy Maxim thread"
>> since the customer fucks up a lot with multiple power supplies, so
>> leaving the positive rail diode off the chip is one less thing for the
>> customer to screw up.
>>
>> Personally, I don't think the snap back Nfet is as good as the diodes to
>> both rails, but it is a matter of customer confidence. Remember, 99.9%
>> of the time, the customer fucks up, but they believe that the
>> manufacturer is the problem. [The manufacturer is selling billions of
>> dollars of compoenents, so obviously some customers know what they are
>> doing.] If the customer is latching up parts due to a sequencing issue
>> with multiple voltages on board, they will design out your part (even
>> though they fucked up), they will bad mouth your company (even though
>> they fucked up), or do a combination of both. Customer shit is like ice
>> cream, while the manufacturer's shit is ..well shit!
>>
>> Thus all sorts of protection go into chips to keep the moronic customer
>> at bay. You have to assume some oaf will short adjacent pins together
>> with a scope probe, so you think about what will happen under those
>> conditions. If you have a power device on chip, you put in some short
>> circuit protection just in case the customer shorts it to a rail, since
>> their clumsiness is of course the manufacturer's fault.
>>
>> You can't believe the amount of work that goes into a power on reset /
>> undervoltage lockout scheme. Some customers are so fucking cheap they
>> design their owner DC/DC rather than use a good controller chip, so you
>> need to be ready for really bad power supply start up issues. Never ever
>> assume power supply power up will be monotonic.
>>
>>
>
>Well, of course we poor benighted customers never realized that our
>problems were so much less important than yours. I apologize deeply and
>promise that I will never, never complain again, even when some poor
>victimized IC house fails to ship the promised wonder-chip for the Nth
>time in a row.
>
>Could I adjust your pillow to make you more comfortable? How about a
>nice cup of tea?
Miso worked for Maxim so understand where his head is.
Reply by Jim Thompson●October 23, 20132013-10-23
On Wed, 23 Oct 2013 16:49:56 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:
>On 10/23/2013 04:27 PM, miso wrote:
>> On 10/20/2013 12:24 PM, John Larkin wrote:
>>> On Sun, 20 Oct 2013 14:07:41 -0400, Phil Hobbs
>>> <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>>
>>>> On 10/20/2013 2:03 PM, John Larkin wrote:
>>>>>
>>>>>
>>>>> Lots of CMOS gates and such are tolerant of pin voltages above Vcc,
>>>>> typically
>>>>> rated for 7 volts max. Does anybody know what sort of input circuits
>>>>> are
>>>>> typically used? Does it behave like a zener? To ground?
>>>>>
>>>>> Some parts specify an allowed input curent in the pullup direction,
>>>>> some don't.
>>>>>
>>>>> I'm designing a really tiny board, and an opamp can potentially
>>>>> overdrive an
>>>>> analog mux, an FSA3157. I can current limit it with a "free"
>>>>> resistor, in a
>>>>> pack, but a diode would take more room.
>>>>
>>>> Overdriving an analogue mux can have a lot of nasty effects without
>>>> blowing up the part, e.g. connecting all the inputs together. How about
>>>> a nice voltage divider to prevent the overdrive in the first place?
>>>>
>>>
>>> I don't mind the mux getting weird, I just don't want to fry it. If I
>>> absolutely
>>> have to add a part, it could be a tiny schottky diode to +5. I just
>>> begrudge
>>> every part on this schematic; the PCB will be under 2 square inches.
>>> I'll be
>>> using a dual opamp and the other half needs to be powered from +12, so
>>> the one
>>> that I'm gain-switching x1/x10 with the mux is +12 too.
>>>
>>> But I've been generally interested in finding out what's inside those
>>> "5 volt
>>> tolerant" cmos parts. The positive direction must have some sort of
>>> zener-y
>>> thing, for ESD.
>>>
>>>
>>
>> ESD in the negative direction is with a body diode. The positive
>> direction depends on the device breaking down, i.e. snap back, and of
>> course surviving. This is very old technology dating back to the early
>> 90s, but not really advertised much until mixed voltages became more
>> common. You may recall I mentioned this in your "Don't buy Maxim thread"
>> since the customer fucks up a lot with multiple power supplies, so
>> leaving the positive rail diode off the chip is one less thing for the
>> customer to screw up.
>>
>> Personally, I don't think the snap back Nfet is as good as the diodes to
>> both rails, but it is a matter of customer confidence. Remember, 99.9%
>> of the time, the customer fucks up, but they believe that the
>> manufacturer is the problem. [The manufacturer is selling billions of
>> dollars of compoenents, so obviously some customers know what they are
>> doing.] If the customer is latching up parts due to a sequencing issue
>> with multiple voltages on board, they will design out your part (even
>> though they fucked up), they will bad mouth your company (even though
>> they fucked up), or do a combination of both. Customer shit is like ice
>> cream, while the manufacturer's shit is ..well shit!
>>
>> Thus all sorts of protection go into chips to keep the moronic customer
>> at bay. You have to assume some oaf will short adjacent pins together
>> with a scope probe, so you think about what will happen under those
>> conditions. If you have a power device on chip, you put in some short
>> circuit protection just in case the customer shorts it to a rail, since
>> their clumsiness is of course the manufacturer's fault.
>>
>> You can't believe the amount of work that goes into a power on reset /
>> undervoltage lockout scheme. Some customers are so fucking cheap they
>> design their owner DC/DC rather than use a good controller chip, so you
>> need to be ready for really bad power supply start up issues. Never ever
>> assume power supply power up will be monotonic.
>>
>>
>
>Well, of course we poor benighted customers never realized that our
>problems were so much less important than yours. I apologize deeply and
>promise that I will never, never complain again, even when some poor
>victimized IC house fails to ship the promised wonder-chip for the Nth
>time in a row.
>
>Could I adjust your pillow to make you more comfortable? How about a
>nice cup of tea?
>
>;)
>
>Cheers
>
>Phil Hobbs
I'd like lemon in mine ;-)
...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 Skype: Contacts Only | |
| Voice:(480)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.
Reply by Phil Hobbs●October 23, 20132013-10-23
On 10/23/2013 04:27 PM, miso wrote:
> On 10/20/2013 12:24 PM, John Larkin wrote:
>> On Sun, 20 Oct 2013 14:07:41 -0400, Phil Hobbs
>> <pcdhSpamMeSenseless@electrooptical.net> wrote:
>>
>>> On 10/20/2013 2:03 PM, John Larkin wrote:
>>>>
>>>>
>>>> Lots of CMOS gates and such are tolerant of pin voltages above Vcc,
>>>> typically
>>>> rated for 7 volts max. Does anybody know what sort of input circuits
>>>> are
>>>> typically used? Does it behave like a zener? To ground?
>>>>
>>>> Some parts specify an allowed input curent in the pullup direction,
>>>> some don't.
>>>>
>>>> I'm designing a really tiny board, and an opamp can potentially
>>>> overdrive an
>>>> analog mux, an FSA3157. I can current limit it with a "free"
>>>> resistor, in a
>>>> pack, but a diode would take more room.
>>>
>>> Overdriving an analogue mux can have a lot of nasty effects without
>>> blowing up the part, e.g. connecting all the inputs together. How about
>>> a nice voltage divider to prevent the overdrive in the first place?
>>>
>>
>> I don't mind the mux getting weird, I just don't want to fry it. If I
>> absolutely
>> have to add a part, it could be a tiny schottky diode to +5. I just
>> begrudge
>> every part on this schematic; the PCB will be under 2 square inches.
>> I'll be
>> using a dual opamp and the other half needs to be powered from +12, so
>> the one
>> that I'm gain-switching x1/x10 with the mux is +12 too.
>>
>> But I've been generally interested in finding out what's inside those
>> "5 volt
>> tolerant" cmos parts. The positive direction must have some sort of
>> zener-y
>> thing, for ESD.
>>
>>
>
> ESD in the negative direction is with a body diode. The positive
> direction depends on the device breaking down, i.e. snap back, and of
> course surviving. This is very old technology dating back to the early
> 90s, but not really advertised much until mixed voltages became more
> common. You may recall I mentioned this in your "Don't buy Maxim thread"
> since the customer fucks up a lot with multiple power supplies, so
> leaving the positive rail diode off the chip is one less thing for the
> customer to screw up.
>
> Personally, I don't think the snap back Nfet is as good as the diodes to
> both rails, but it is a matter of customer confidence. Remember, 99.9%
> of the time, the customer fucks up, but they believe that the
> manufacturer is the problem. [The manufacturer is selling billions of
> dollars of compoenents, so obviously some customers know what they are
> doing.] If the customer is latching up parts due to a sequencing issue
> with multiple voltages on board, they will design out your part (even
> though they fucked up), they will bad mouth your company (even though
> they fucked up), or do a combination of both. Customer shit is like ice
> cream, while the manufacturer's shit is ..well shit!
>
> Thus all sorts of protection go into chips to keep the moronic customer
> at bay. You have to assume some oaf will short adjacent pins together
> with a scope probe, so you think about what will happen under those
> conditions. If you have a power device on chip, you put in some short
> circuit protection just in case the customer shorts it to a rail, since
> their clumsiness is of course the manufacturer's fault.
>
> You can't believe the amount of work that goes into a power on reset /
> undervoltage lockout scheme. Some customers are so fucking cheap they
> design their owner DC/DC rather than use a good controller chip, so you
> need to be ready for really bad power supply start up issues. Never ever
> assume power supply power up will be monotonic.
>
>
Well, of course we poor benighted customers never realized that our
problems were so much less important than yours. I apologize deeply and
promise that I will never, never complain again, even when some poor
victimized IC house fails to ship the promised wonder-chip for the Nth
time in a row.
Could I adjust your pillow to make you more comfortable? How about a
nice cup of tea?
;)
Cheers
Phil Hobbs
--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics
160 North State Road #203
Briarcliff Manor NY 10510
hobbs at electrooptical dot net
http://electrooptical.net
Reply by miso●October 23, 20132013-10-23
On 10/20/2013 12:24 PM, John Larkin wrote:
> On Sun, 20 Oct 2013 14:07:41 -0400, Phil Hobbs
> <pcdhSpamMeSenseless@electrooptical.net> wrote:
>
>> On 10/20/2013 2:03 PM, John Larkin wrote:
>>>
>>>
>>> Lots of CMOS gates and such are tolerant of pin voltages above Vcc, typically
>>> rated for 7 volts max. Does anybody know what sort of input circuits are
>>> typically used? Does it behave like a zener? To ground?
>>>
>>> Some parts specify an allowed input curent in the pullup direction, some don't.
>>>
>>> I'm designing a really tiny board, and an opamp can potentially overdrive an
>>> analog mux, an FSA3157. I can current limit it with a "free" resistor, in a
>>> pack, but a diode would take more room.
>>
>> Overdriving an analogue mux can have a lot of nasty effects without
>> blowing up the part, e.g. connecting all the inputs together. How about
>> a nice voltage divider to prevent the overdrive in the first place?
>>
>
> I don't mind the mux getting weird, I just don't want to fry it. If I absolutely
> have to add a part, it could be a tiny schottky diode to +5. I just begrudge
> every part on this schematic; the PCB will be under 2 square inches. I'll be
> using a dual opamp and the other half needs to be powered from +12, so the one
> that I'm gain-switching x1/x10 with the mux is +12 too.
>
> But I've been generally interested in finding out what's inside those "5 volt
> tolerant" cmos parts. The positive direction must have some sort of zener-y
> thing, for ESD.
>
>
ESD in the negative direction is with a body diode. The positive
direction depends on the device breaking down, i.e. snap back, and of
course surviving. This is very old technology dating back to the early
90s, but not really advertised much until mixed voltages became more
common. You may recall I mentioned this in your "Don't buy Maxim thread"
since the customer fucks up a lot with multiple power supplies, so
leaving the positive rail diode off the chip is one less thing for the
customer to screw up.
Personally, I don't think the snap back Nfet is as good as the diodes to
both rails, but it is a matter of customer confidence. Remember, 99.9%
of the time, the customer fucks up, but they believe that the
manufacturer is the problem. [The manufacturer is selling billions of
dollars of compoenents, so obviously some customers know what they are
doing.] If the customer is latching up parts due to a sequencing issue
with multiple voltages on board, they will design out your part (even
though they fucked up), they will bad mouth your company (even though
they fucked up), or do a combination of both. Customer shit is like ice
cream, while the manufacturer's shit is ..well shit!
Thus all sorts of protection go into chips to keep the moronic customer
at bay. You have to assume some oaf will short adjacent pins together
with a scope probe, so you think about what will happen under those
conditions. If you have a power device on chip, you put in some short
circuit protection just in case the customer shorts it to a rail, since
their clumsiness is of course the manufacturer's fault.
You can't believe the amount of work that goes into a power on reset /
undervoltage lockout scheme. Some customers are so fucking cheap they
design their owner DC/DC rather than use a good controller chip, so you
need to be ready for really bad power supply start up issues. Never ever
assume power supply power up will be monotonic.
Reply by John Larkin●October 22, 20132013-10-22
On Mon, 21 Oct 2013 15:34:13 -0400, Spehro Pefhany
<speffSNIP@interlogDOTyou.knowwhat> wrote:
>On Mon, 21 Oct 2013 11:33:03 -0700 (PDT),
>bloggs.fredbloggs.fred@gmail.com wrote:
>
>>On Sunday, October 20, 2013 2:03:02 PM UTC-4, John Larkin wrote:
>>> Lots of CMOS gates and such are tolerant of pin voltages above Vcc, typically
>>>
>>> rated for 7 volts max. Does anybody know what sort of input circuits are
>>>
>>> typically used? Does it behave like a zener? To ground?
>>>
>>>
>>>
>>> Some parts specify an allowed input curent in the pullup direction, some don't.
>>>
>>>
>>>
>>> I'm designing a really tiny board, and an opamp can potentially overdrive an
>>>
>>> analog mux, an FSA3157. I can current limit it with a "free" resistor, in a
>>>
>>> pack, but a diode would take more room.
>>>
>>>
>>>
>>>
>>>
>>> --
>>>
>>>
>>>
>>> John Larkin Highland Technology Inc
>>>
>>> www.highlandtechnology.com jlarkin at highlandtechnology dot com
>>>
>>>
>>>
>>> Precision electronic instrumentation
>>>
>>> Picosecond-resolution Digital Delay and Pulse generators
>>>
>>> Custom timing and laser controllers
>>>
>>> Photonics and fiberoptic TTL data links
>>>
>>> VME analog, thermocouple, LVDT, synchro, tachometer
>>>
>>> Multichannel arbitrary waveform generators
>>
>>Say what you want about Maxim, but they do have this incredible beyond-the-rails analog switch technology:
>>http://www.maximintegrated.com/datasheet/index.mvp/id/7503
>
>Three internal 35MHz charge pumps. Their guaranteed max leakage is
>pretty high, but they'd be great for low-distortion switching of
>reasonably low impedance/low voltage circuits. Not so great for
>precision.
>
>Hey, maybe John can use one of these as a bias generator for a PD--
>they should output +35 and -27 from a 3-5V source. If they can
>actually be purchased.
Some sort of IC bias generator would be cute, if it didn't violate the Never Buy
Maxim rule.
As an analog switch, that has way too much personality. I'd just as soon use an
opto-SSR.
In my current project, I'll keep the pd bias supply out of the optical head. If
I did put it on the same board with the photodiode, it would be a PV isolator or
a soft, slow, low KHz charge pump.
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators