Reply by Lasse Langwadt Christensen●April 11, 20232023-04-11
tirsdag den 11. april 2023 kl. 19.59.12 UTC+2 skrev Ricky:
> On Tuesday, April 11, 2023 at 11:16:03 AM UTC-4, Lasse Langwadt Christensen wrote:
> > tirsdag den 11. april 2023 kl. 17.01.21 UTC+2 skrev John Larkin:
> > > On Tue, 11 Apr 2023 07:34:25 -0700 (PDT), Lasse Langwadt Christensen
> > > <lang...@fonz.dk> wrote:
> > >
> > > >tirsdag den 11. april 2023 kl. 16.27.42 UTC+2 skrev John Larkin:
> > > >> On Tue, 11 Apr 2023 14:30:47 +0100, Clive Arthur
> > > >> <cl...@nowaytoday.co.uk> wrote:
> > > >>
> > > >> >On 11/04/2023 12:18, Clive Arthur wrote:
> > > >> >> On 09/04/2023 02:22, Ricky wrote:
> > > >> >>> I'm tired of digging around looking for an optimal chip to level shift
> > > >> >>> a pair of signals from 5V TTL (an RS-422 transceiver) to 3.3V CMOS.
> > > >> >>> There are too many devices, and no small number of them are impacted
> > > >> >>> by the semiconductor shortage. But resistors are pretty available, as
> > > >> >>> are Zener diodes.
> > > >> >>>
> > > >> >>> So, I could use a pair of resistors to simply divide the TTL output
> > > >> >>> voltage to something safe for CMOS. But there's not much wiggle room,
> > > >> >>> given that the TTL output is not well specified. If the current
> > > >> >>> through the divider is minimized, this slows the signal edges. So a
> > > >> >>> mA or two needs to be flowing when the output is high, but not much
> > > >> >>> more, because of the power budget.
> > > >> >>>
> > > >> >>> Another option is to use a Zener diode. Most Zeners are specified at
> > > >> >>> 5 to 20 mA of current. Running at 1 mA is into the knee as best I can
> > > >> >>> tell, not desirable, resulting in poor regulation.
> > > >> >>>
> > > >> >>> I'm thinking with the variation in output voltage from the TTL output,
> > > >> >>> the Zener might not do much better than resistors.
> > > >> >>>
> > > >> >>> Any thoughts?
> > > >> >>>
> > > >> >>
> > > >> >> Series diode or two, pull down resistor.
> > > >> >>
> > > >> >
> > > >> >Or single 1.8V series zener, if your 0V doesn't mind being ~0.7V or so.
> > > >> What's the zener drop?
> > > >>
> > > >> What's the zener current?
> > > >>
> > > >> Where can I buy a 1.8v zener?
> > > >
> > > >https://www.mouser.dk/ProductDetail/Nexperia/BZX8850S-C1V8-QYL?qs=7D1LtPJG0i3BPC7qWB9brw%3D%3D
> > > What will the zener voltage drop be in the proposed circuit?
> > does it matter? if a simple resistor is somehow not good enough there's no point in not using a 5V tolerant buffer
> If it makes anyone feel better, I did find the spec on the part involved. They allow 10 mA into the protection diodes for a pin and 120 mA total for a bank. But, since they have no app note or any further info in the data sheet, there is no specific exception given to the max Vin on the I/Os, beyond Vcc + 0.3V. There's also the power consumption issue. A resistor of any given value would have a power drain amount that varies 3 to 1 or more.
>
with the series resistor the current taken from the 5V ends up in the 3.3V so if 3.3V is made with a linear regulator from 5V it is not wasted
Reply by Ricky●April 11, 20232023-04-11
On Tuesday, April 11, 2023 at 11:16:03 AM UTC-4, Lasse Langwadt Christensen wrote:
> tirsdag den 11. april 2023 kl. 17.01.21 UTC+2 skrev John Larkin:
> > On Tue, 11 Apr 2023 07:34:25 -0700 (PDT), Lasse Langwadt Christensen
> > <lang...@fonz.dk> wrote:
> >
> > >tirsdag den 11. april 2023 kl. 16.27.42 UTC+2 skrev John Larkin:
> > >> On Tue, 11 Apr 2023 14:30:47 +0100, Clive Arthur
> > >> <cl...@nowaytoday.co.uk> wrote:
> > >>
> > >> >On 11/04/2023 12:18, Clive Arthur wrote:
> > >> >> On 09/04/2023 02:22, Ricky wrote:
> > >> >>> I'm tired of digging around looking for an optimal chip to level shift
> > >> >>> a pair of signals from 5V TTL (an RS-422 transceiver) to 3.3V CMOS.
> > >> >>> There are too many devices, and no small number of them are impacted
> > >> >>> by the semiconductor shortage. But resistors are pretty available, as
> > >> >>> are Zener diodes.
> > >> >>>
> > >> >>> So, I could use a pair of resistors to simply divide the TTL output
> > >> >>> voltage to something safe for CMOS. But there's not much wiggle room,
> > >> >>> given that the TTL output is not well specified. If the current
> > >> >>> through the divider is minimized, this slows the signal edges. So a
> > >> >>> mA or two needs to be flowing when the output is high, but not much
> > >> >>> more, because of the power budget.
> > >> >>>
> > >> >>> Another option is to use a Zener diode. Most Zeners are specified at
> > >> >>> 5 to 20 mA of current. Running at 1 mA is into the knee as best I can
> > >> >>> tell, not desirable, resulting in poor regulation.
> > >> >>>
> > >> >>> I'm thinking with the variation in output voltage from the TTL output,
> > >> >>> the Zener might not do much better than resistors.
> > >> >>>
> > >> >>> Any thoughts?
> > >> >>>
> > >> >>
> > >> >> Series diode or two, pull down resistor.
> > >> >>
> > >> >
> > >> >Or single 1.8V series zener, if your 0V doesn't mind being ~0.7V or so.
> > >> What's the zener drop?
> > >>
> > >> What's the zener current?
> > >>
> > >> Where can I buy a 1.8v zener?
> > >
> > >https://www.mouser.dk/ProductDetail/Nexperia/BZX8850S-C1V8-QYL?qs=7D1LtPJG0i3BPC7qWB9brw%3D%3D
> > What will the zener voltage drop be in the proposed circuit?
> does it matter? if a simple resistor is somehow not good enough there's no point in not using a 5V tolerant buffer
If it makes anyone feel better, I did find the spec on the part involved. They allow 10 mA into the protection diodes for a pin and 120 mA total for a bank. But, since they have no app note or any further info in the data sheet, there is no specific exception given to the max Vin on the I/Os, beyond Vcc + 0.3V. There's also the power consumption issue. A resistor of any given value would have a power drain amount that varies 3 to 1 or more.
The two resistor approach has better potential. The ratio of resistors would be something like 3:1, drawing limited current when high, and a low impedance for speed. But even a pair of 0402 parts is as large as the dual inverter.
At this point, I'm good with the dual inverter with specified 5V tolerant I/Os. I'm not interested in exploring the analog properties of a large digital device.
--
Rick C.
-+-- Get 1,000 miles of free Supercharging
-+-- Tesla referral code - https://ts.la/richard11209
Reply by Lasse Langwadt Christensen●April 11, 20232023-04-11
tirsdag den 11. april 2023 kl. 17.38.41 UTC+2 skrev Ralph Mowery:
> In article <gfra3idpgu69fmb7r...@4ax.com>,
> jjla...@highlandtechnology.com says...
> >
> > >Or single 1.8V series zener, if your 0V doesn't mind being ~0.7V or so.
> >
> > What's the zener drop?
> >
> > What's the zener current?
> >
> > Where can I buy a 1.8v zener?
> >
> >
> >
> Depending on the current some LEDs are in that range. You get a zener
> and a light to boot.
an LED is missing the diode in the other direction
Reply by Ralph Mowery●April 11, 20232023-04-11
In article <gfra3idpgu69fmb7rfbv6et3dp9jphu5cm@4ax.com>,
jjlarkin@highlandtechnology.com says...
>
> >Or single 1.8V series zener, if your 0V doesn't mind being ~0.7V or so.
>
> What's the zener drop?
>
> What's the zener current?
>
> Where can I buy a 1.8v zener?
>
>
>
Depending on the current some LEDs are in that range. You get a zener
and a light to boot.
Reply by Lasse Langwadt Christensen●April 11, 20232023-04-11
tirsdag den 11. april 2023 kl. 17.01.21 UTC+2 skrev John Larkin:
> On Tue, 11 Apr 2023 07:34:25 -0700 (PDT), Lasse Langwadt Christensen
> <lang...@fonz.dk> wrote:
>
> >tirsdag den 11. april 2023 kl. 16.27.42 UTC+2 skrev John Larkin:
> >> On Tue, 11 Apr 2023 14:30:47 +0100, Clive Arthur
> >> <cl...@nowaytoday.co.uk> wrote:
> >>
> >> >On 11/04/2023 12:18, Clive Arthur wrote:
> >> >> On 09/04/2023 02:22, Ricky wrote:
> >> >>> I'm tired of digging around looking for an optimal chip to level shift
> >> >>> a pair of signals from 5V TTL (an RS-422 transceiver) to 3.3V CMOS.
> >> >>> There are too many devices, and no small number of them are impacted
> >> >>> by the semiconductor shortage. But resistors are pretty available, as
> >> >>> are Zener diodes.
> >> >>>
> >> >>> So, I could use a pair of resistors to simply divide the TTL output
> >> >>> voltage to something safe for CMOS. But there's not much wiggle room,
> >> >>> given that the TTL output is not well specified. If the current
> >> >>> through the divider is minimized, this slows the signal edges. So a
> >> >>> mA or two needs to be flowing when the output is high, but not much
> >> >>> more, because of the power budget.
> >> >>>
> >> >>> Another option is to use a Zener diode. Most Zeners are specified at
> >> >>> 5 to 20 mA of current. Running at 1 mA is into the knee as best I can
> >> >>> tell, not desirable, resulting in poor regulation.
> >> >>>
> >> >>> I'm thinking with the variation in output voltage from the TTL output,
> >> >>> the Zener might not do much better than resistors.
> >> >>>
> >> >>> Any thoughts?
> >> >>>
> >> >>
> >> >> Series diode or two, pull down resistor.
> >> >>
> >> >
> >> >Or single 1.8V series zener, if your 0V doesn't mind being ~0.7V or so.
> >> What's the zener drop?
> >>
> >> What's the zener current?
> >>
> >> Where can I buy a 1.8v zener?
> >
> >https://www.mouser.dk/ProductDetail/Nexperia/BZX8850S-C1V8-QYL?qs=7D1LtPJG0i3BPC7qWB9brw%3D%3D
> What will the zener voltage drop be in the proposed circuit?
does it matter? if a simple resistor is somehow not good enough there's no point in not using a 5V tolerant buffer
Reply by John Larkin●April 11, 20232023-04-11
On Tue, 11 Apr 2023 07:34:25 -0700 (PDT), Lasse Langwadt Christensen
<langwadt@fonz.dk> wrote:
>tirsdag den 11. april 2023 kl. 16.27.42 UTC+2 skrev John Larkin:
>> On Tue, 11 Apr 2023 14:30:47 +0100, Clive Arthur
>> <cl...@nowaytoday.co.uk> wrote:
>>
>> >On 11/04/2023 12:18, Clive Arthur wrote:
>> >> On 09/04/2023 02:22, Ricky wrote:
>> >>> I'm tired of digging around looking for an optimal chip to level shift
>> >>> a pair of signals from 5V TTL (an RS-422 transceiver) to 3.3V CMOS.
>> >>> There are too many devices, and no small number of them are impacted
>> >>> by the semiconductor shortage. But resistors are pretty available, as
>> >>> are Zener diodes.
>> >>>
>> >>> So, I could use a pair of resistors to simply divide the TTL output
>> >>> voltage to something safe for CMOS. But there's not much wiggle room,
>> >>> given that the TTL output is not well specified. If the current
>> >>> through the divider is minimized, this slows the signal edges. So a
>> >>> mA or two needs to be flowing when the output is high, but not much
>> >>> more, because of the power budget.
>> >>>
>> >>> Another option is to use a Zener diode. Most Zeners are specified at
>> >>> 5 to 20 mA of current. Running at 1 mA is into the knee as best I can
>> >>> tell, not desirable, resulting in poor regulation.
>> >>>
>> >>> I'm thinking with the variation in output voltage from the TTL output,
>> >>> the Zener might not do much better than resistors.
>> >>>
>> >>> Any thoughts?
>> >>>
>> >>
>> >> Series diode or two, pull down resistor.
>> >>
>> >
>> >Or single 1.8V series zener, if your 0V doesn't mind being ~0.7V or so.
>> What's the zener drop?
>>
>> What's the zener current?
>>
>> Where can I buy a 1.8v zener?
>
>https://www.mouser.dk/ProductDetail/Nexperia/BZX8850S-C1V8-QYL?qs=7D1LtPJG0i3BPC7qWB9brw%3D%3D
What will the zener voltage drop be in the proposed circuit?
Reply by Ricky●April 11, 20232023-04-11
On Tuesday, April 11, 2023 at 10:34:29 AM UTC-4, Lasse Langwadt Christensen wrote:
> tirsdag den 11. april 2023 kl. 16.27.42 UTC+2 skrev John Larkin:
> > On Tue, 11 Apr 2023 14:30:47 +0100, Clive Arthur
> > <cl...@nowaytoday.co.uk> wrote:
> >
> > >On 11/04/2023 12:18, Clive Arthur wrote:
> > >> On 09/04/2023 02:22, Ricky wrote:
> > >>> I'm tired of digging around looking for an optimal chip to level shift
> > >>> a pair of signals from 5V TTL (an RS-422 transceiver) to 3.3V CMOS.
> > >>> There are too many devices, and no small number of them are impacted
> > >>> by the semiconductor shortage. But resistors are pretty available, as
> > >>> are Zener diodes.
> > >>>
> > >>> So, I could use a pair of resistors to simply divide the TTL output
> > >>> voltage to something safe for CMOS. But there's not much wiggle room,
> > >>> given that the TTL output is not well specified. If the current
> > >>> through the divider is minimized, this slows the signal edges. So a
> > >>> mA or two needs to be flowing when the output is high, but not much
> > >>> more, because of the power budget.
> > >>>
> > >>> Another option is to use a Zener diode. Most Zeners are specified at
> > >>> 5 to 20 mA of current. Running at 1 mA is into the knee as best I can
> > >>> tell, not desirable, resulting in poor regulation.
> > >>>
> > >>> I'm thinking with the variation in output voltage from the TTL output,
> > >>> the Zener might not do much better than resistors.
> > >>>
> > >>> Any thoughts?
> > >>>
> > >>
> > >> Series diode or two, pull down resistor.
> > >>
> > >
> > >Or single 1.8V series zener, if your 0V doesn't mind being ~0.7V or so.
> > What's the zener drop?
> >
> > What's the zener current?
> >
> > Where can I buy a 1.8v zener?
> https://www.mouser.dk/ProductDetail/Nexperia/BZX8850S-C1V8-QYL?qs=7D1LtPJG0i3BPC7qWB9brw%3D%3D
I actually did that once. I had a board in production where I had forgotten about the voltage difference. I added a zener in series and let it play against the protection diode. It was a really ugly solution, but could be done without a board spin. Maybe that's why I don't like the series resistor idea. This is all too kludgy for me. Is 200 ohms enough? Is 600 ohms too many?
The logic parts are tiny and easy to buy. 5V tolerant regardless of Vcc. Yeah, that works great for me!
--
Rick C.
--++ Get 1,000 miles of free Supercharging
--++ Tesla referral code - https://ts.la/richard11209
Reply by Lasse Langwadt Christensen●April 11, 20232023-04-11
tirsdag den 11. april 2023 kl. 16.27.42 UTC+2 skrev John Larkin:
> On Tue, 11 Apr 2023 14:30:47 +0100, Clive Arthur
> <cl...@nowaytoday.co.uk> wrote:
>
> >On 11/04/2023 12:18, Clive Arthur wrote:
> >> On 09/04/2023 02:22, Ricky wrote:
> >>> I'm tired of digging around looking for an optimal chip to level shift
> >>> a pair of signals from 5V TTL (an RS-422 transceiver) to 3.3V CMOS.
> >>> There are too many devices, and no small number of them are impacted
> >>> by the semiconductor shortage. But resistors are pretty available, as
> >>> are Zener diodes.
> >>>
> >>> So, I could use a pair of resistors to simply divide the TTL output
> >>> voltage to something safe for CMOS. But there's not much wiggle room,
> >>> given that the TTL output is not well specified. If the current
> >>> through the divider is minimized, this slows the signal edges. So a
> >>> mA or two needs to be flowing when the output is high, but not much
> >>> more, because of the power budget.
> >>>
> >>> Another option is to use a Zener diode. Most Zeners are specified at
> >>> 5 to 20 mA of current. Running at 1 mA is into the knee as best I can
> >>> tell, not desirable, resulting in poor regulation.
> >>>
> >>> I'm thinking with the variation in output voltage from the TTL output,
> >>> the Zener might not do much better than resistors.
> >>>
> >>> Any thoughts?
> >>>
> >>
> >> Series diode or two, pull down resistor.
> >>
> >
> >Or single 1.8V series zener, if your 0V doesn't mind being ~0.7V or so.
> What's the zener drop?
>
> What's the zener current?
>
> Where can I buy a 1.8v zener?
On Tue, 11 Apr 2023 14:30:47 +0100, Clive Arthur
<clive@nowaytoday.co.uk> wrote:
>On 11/04/2023 12:18, Clive Arthur wrote:
>> On 09/04/2023 02:22, Ricky wrote:
>>> I'm tired of digging around looking for an optimal chip to level shift
>>> a pair of signals from 5V TTL (an RS-422 transceiver) to 3.3V CMOS.
>>> There are too many devices, and no small number of them are impacted
>>> by the semiconductor shortage.� But resistors are pretty available, as
>>> are Zener diodes.
>>>
>>> So, I could use a pair of resistors to simply divide the TTL output
>>> voltage to something safe for CMOS.� But there's not much wiggle room,
>>> given that the TTL output is not well specified.� If the current
>>> through the divider is minimized, this slows the signal edges.� So a
>>> mA or two needs to be flowing when the output is high, but not much
>>> more, because of the power budget.
>>>
>>> Another option is to use a Zener diode.� Most Zeners are specified at
>>> 5 to 20 mA of current. Running at 1 mA is into the knee as best I can
>>> tell, not desirable, resulting in poor regulation.
>>>
>>> I'm thinking with the variation in output voltage from the TTL output,
>>> the Zener might not do much better than resistors.
>>>
>>> Any thoughts?
>>>
>>
>> Series diode or two, pull down resistor.
>>
>
>Or single 1.8V series zener, if your 0V doesn't mind being ~0.7V or so.
What's the zener drop?
What's the zener current?
Where can I buy a 1.8v zener?
Reply by Ricky●April 11, 20232023-04-11
On Tuesday, April 11, 2023 at 10:12:35 AM UTC-4, Clive Arthur wrote:
> On 11/04/2023 14:53, Ricky wrote:
> > On Tuesday, April 11, 2023 at 9:30:56 AM UTC-4, Clive Arthur wrote:
> >> On 11/04/2023 12:18, Clive Arthur wrote:
> >>> On 09/04/2023 02:22, Ricky wrote:
> >>>> I'm tired of digging around looking for an optimal chip to level shift
> >>>> a pair of signals from 5V TTL (an RS-422 transceiver) to 3.3V CMOS.
> >>>> There are too many devices, and no small number of them are impacted
> >>>> by the semiconductor shortage. But resistors are pretty available, as
> >>>> are Zener diodes.
> >>>>
> >>>> So, I could use a pair of resistors to simply divide the TTL output
> >>>> voltage to something safe for CMOS. But there's not much wiggle room,
> >>>> given that the TTL output is not well specified. If the current
> >>>> through the divider is minimized, this slows the signal edges. So a
> >>>> mA or two needs to be flowing when the output is high, but not much
> >>>> more, because of the power budget.
> >>>>
> >>>> Another option is to use a Zener diode. Most Zeners are specified at
> >>>> 5 to 20 mA of current. Running at 1 mA is into the knee as best I can
> >>>> tell, not desirable, resulting in poor regulation.
> >>>>
> >>>> I'm thinking with the variation in output voltage from the TTL output,
> >>>> the Zener might not do much better than resistors.
> >>>>
> >>>> Any thoughts?
> >>>>
> >>>
> >>> Series diode or two, pull down resistor.
> >>>
> >> Or single 1.8V series zener, if your 0V doesn't mind being ~0.7V or so.
> >
> > You mean +0.7V, no?
> >
> Yeah, bad habit, I tend to use the tilde to mean approx.
Sorry. I'm having a poor vision morning and thought it was a -. Tilde I understand.
Thanks for your suggestions.
--
Rick C.
--+- Get 1,000 miles of free Supercharging
--+- Tesla referral code - https://ts.la/richard11209