> On Fri, 24 Nov 2017 08:28:06 -0800, Joerg <news@analogconsultants.com>
> wrote:
>
>> On 2017-11-24 08:18, John Devereux wrote:
>>> John Larkin <jjlarkin@highlandtechnology.com> writes:
>>>
>>>> On Thu, 23 Nov 2017 07:54:57 -0800, Joerg <news@analogconsultants.com>
>>>> wrote:
>>>>
>>>>> On 2017-11-22 22:22, Jasen Betts wrote:
>>>>>> On 2017-11-21, Joerg <news@analogconsultants.com> wrote:
>>>>>>> On 2017-11-21 11:35, Winfield Hill wrote:
>>>>>>>> John Larkin wrote...
>>>>>>>>>
>>>>>>>>> The ancient LM324 has PNP input transistors, and the
>>>>>>>>> safe input voltages can go to +32 irrespective of V+.
>>>>>>>>
>>>>>>>
>>>>>>> The datasheet says "Don't do that" though. Looking at the innards on
>>>>>>> page 4 it seems you'd hit a diode path to V+:
>>>>>>>
>>>>>>> http://www.st.com/content/ccc/resource/technical/document/datasheet/bd/fc/46/43/26/8f/40/7f/CD00001046.pdf/files/CD00001046.pdf/jcr:content/translations/en.CD00001046.pdf
>>>>>>
>>>>>> use the TI part.
>>>>>>
>>>>>> http://www.ti.com/lit/ds/snosc16d/snosc16d.pdf
>>>>>>
>>>>>
>>>>> For hobby or small runs it's ok. With contract assemblers it can be
>>>>> dicey.
>>>>
>>>> NEVER allow a CM to substitute parts!
>>>
>>> Yes, quite, I am amazed that Jeorg allows this to happen actually given
>>> his previous postings here. (Regarding documentation control,
>>> certifications, procedures, etc etc).
>>>
>>
>> As I wrote I didn't allow it and neither did my client. The CM folks
>> took it upon themselves to "cost-optimize". Which is why they had to eat
>> the cost for an expedited re-run and fast air-shipment.
>>
>>
>>> Also it seems rather unfair on the original manufacturer of the part. As
>>> soon as someone releases a clone, only hobbyists should then use the
>>> original?
>>>
>>
>> In this case it isn't a clone and if one knows about it then IMO one
>> needs to point out the risk.
>>
>> The other question in this case is that even for the versions that allow
>> an unusually high reverse Vbe there is no spec in the datasheets about
>> the maximum leakage current that will flow, AFAIK. Li-Ion protection is
>> a pretty serious matter, not something to take lightly.
>
> One amp-hour divided by one microampere is a million hours.
>
Leslie was looking for nanoamps, no idea why. Even a few tens of uA
should be ok with 18650 and similar cells. Milliamps, not so much. Had
my comeuppance there with bicycle lights where a standby LED is always
lit. Silly design and draws 5-10mA. When I used the road bike for
several weeks and then wanted to ride my mountain bike I hopped on,
click, no lights.
I don't know what a LM324 draws with the input 25V or so above V+. She
could find out.
If it was my projects I'd first look for complete battery management ICs
with ship mode or similar low power modes. Less space, less parts,
approved design, cheap, no hassle.
--
Regards, Joerg
http://www.analogconsultants.com/
Reply by John Larkin●November 26, 20172017-11-26
On Fri, 24 Nov 2017 08:28:06 -0800, Joerg <news@analogconsultants.com>
wrote:
>On 2017-11-24 08:18, John Devereux wrote:
>> John Larkin <jjlarkin@highlandtechnology.com> writes:
>>
>>> On Thu, 23 Nov 2017 07:54:57 -0800, Joerg <news@analogconsultants.com>
>>> wrote:
>>>
>>>> On 2017-11-22 22:22, Jasen Betts wrote:
>>>>> On 2017-11-21, Joerg <news@analogconsultants.com> wrote:
>>>>>> On 2017-11-21 11:35, Winfield Hill wrote:
>>>>>>> John Larkin wrote...
>>>>>>>>
>>>>>>>> The ancient LM324 has PNP input transistors, and the
>>>>>>>> safe input voltages can go to +32 irrespective of V+.
>>>>>>>
>>>>>>
>>>>>> The datasheet says "Don't do that" though. Looking at the innards on
>>>>>> page 4 it seems you'd hit a diode path to V+:
>>>>>>
>>>>>> http://www.st.com/content/ccc/resource/technical/document/datasheet/bd/fc/46/43/26/8f/40/7f/CD00001046.pdf/files/CD00001046.pdf/jcr:content/translations/en.CD00001046.pdf
>>>>>
>>>>> use the TI part.
>>>>>
>>>>> http://www.ti.com/lit/ds/snosc16d/snosc16d.pdf
>>>>>
>>>>
>>>> For hobby or small runs it's ok. With contract assemblers it can be
>>>> dicey.
>>>
>>> NEVER allow a CM to substitute parts!
>>
>> Yes, quite, I am amazed that Jeorg allows this to happen actually given
>> his previous postings here. (Regarding documentation control,
>> certifications, procedures, etc etc).
>>
>
>As I wrote I didn't allow it and neither did my client. The CM folks
>took it upon themselves to "cost-optimize". Which is why they had to eat
>the cost for an expedited re-run and fast air-shipment.
>
>
>> Also it seems rather unfair on the original manufacturer of the part. As
>> soon as someone releases a clone, only hobbyists should then use the
>> original?
>>
>
>In this case it isn't a clone and if one knows about it then IMO one
>needs to point out the risk.
>
>The other question in this case is that even for the versions that allow
>an unusually high reverse Vbe there is no spec in the datasheets about
>the maximum leakage current that will flow, AFAIK. Li-Ion protection is
>a pretty serious matter, not something to take lightly.
One amp-hour divided by one microampere is a million hours.
--
John Larkin Highland Technology, Inc
lunatic fringe electronics
Reply by ●November 26, 20172017-11-26
On Friday, November 24, 2017 at 5:28:49 PM UTC-5, Leslie Rhorer wrote:
> > >
> > > Am I wrong? If not, is there a different choice of high input impedance op amp that can withstand up to 25V when power is shut off?
> >
> > You should buy a battery management IC, some of them handle lots of cells, and this is especially true of Li batteries. There are physics based considerations that dwarf the triviality of the electronics interconnection topology.
>
> I am using an off-the shelf LiPo battery monitor specifically designed to monitor LiPo batteries during use. These are extremely common in the Radio Control hobby sector, and are readily available at very low cost. Below is an example. The only problem is these are designed to be manually disconnected when not in use. The reed relays work perfectly, but are too large to be able to shrink the design down to a desirable size for a lower power version of the light. The 100W version uses a battery that is nearly 4" x 4" x 12", so a smaller board is not of any value, but I would like to create a few 20W lights with a smaller footprint and less weight.
>
> https://www.ebay.com/itm/5pcs-Lipo-Alarm-Low-Voltage-1-8S-Battery-Buzzer-RC-Checker-Tester-LED-Indicator/263108054278?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2060353.m2749.l2649
Looks like that one is quite popular. One forum convo talked about killing the current draw by opening the GND lead in the off state, all the inputs remain connected. That might be worth exploring.
Reply by Joerg●November 26, 20172017-11-26
On 2017-11-25 21:28, John Larkin wrote:
> On Sat, 25 Nov 2017 17:28:49 -0800 (PST), tabbypurr@gmail.com wrote:
>
>> On Saturday, 25 November 2017 00:09:27 UTC, John Larkin wrote:
>>> On Fri, 24 Nov 2017 14:06:48 -0800 (PST), Leslie Rhorer
>>> <rhorerles@gmail.com> wrote:
>>>> On Friday, November 24, 2017 at 3:48:01 PM UTC-6, John Larkin
>>>> wrote:
>>>>> I don't know of any problems with the original LM324.
>>>>>
>>>>> Except the ghastly crossover distortion.
>>>>>
>>>>> And the pitiful slew rate.
>>>>>
>>>>> And the wimpy drive.
>>>>>
>>>>> And the horrible stuff that happens if any of the inputs go
>>>>> below ground.
>>>>
>>>> None of which matter in this situation, which is one reason why
>>>> your original suggestion is definitely the front runner. I've
>>>> ordered a couple of Op Amps, including a 324, from Mouser for
>>>> testing. They should arrive Monday.
>>>
>>> One other quirk: if one of the four opamps rails, it can mess up
>>> the other three. Shared current sources. I don't remember just
>>> how bad that is.
>>
>> I love the 324, it's cheap & very useful. Crossover is easily
>> avoided. Don't buy Motorola or bias the output.
>>
>> But I do wish there were even cheaper opamps with even more corners
>> cut. There must be plenty of uses for opamps with major
>> compromises, eg poor gain. Trouble is things start expensive and
>> drop in price, which would make cruder opamps uncompetitive in the
>> early days. And who wants to sell an even cheaper chip when they
>> can introduce a high ticket one.
>>
>>
>> NT
>
> The Diodes Inc version of the LM324 is 6.5 cents, Digikey price by
> the reel. Under 2 cents per opamp.
>
> What does Digikey pay? 5 cents? Why does anybody bother to make and
> package a quad opamp for 5 cents?
>
Probably less. You can buy the AS324 for around 4 cents.
Oh yeah. And people with a good meter to match tubes or transistors
could still make a buck.
--
Regards, Joerg
http://www.analogconsultants.com/
Reply by John Larkin●November 26, 20172017-11-26
On Sat, 25 Nov 2017 17:28:49 -0800 (PST), tabbypurr@gmail.com wrote:
>On Saturday, 25 November 2017 00:09:27 UTC, John Larkin wrote:
>> On Fri, 24 Nov 2017 14:06:48 -0800 (PST), Leslie Rhorer
>> <rhorerles@gmail.com> wrote:
>> >On Friday, November 24, 2017 at 3:48:01 PM UTC-6, John Larkin wrote:
>> >> I don't know of any problems with the original LM324.
>> >>
>> >> Except the ghastly crossover distortion.
>> >>
>> >> And the pitiful slew rate.
>> >>
>> >> And the wimpy drive.
>> >>
>> >> And the horrible stuff that happens if any of the inputs go below
>> >> ground.
>> >
>> >None of which matter in this situation, which is one reason why your original suggestion is definitely the front runner. I've ordered a couple of Op Amps, including a 324, from Mouser for testing. They should arrive Monday.
>>
>> One other quirk: if one of the four opamps rails, it can mess up the
>> other three. Shared current sources. I don't remember just how bad
>> that is.
>
>I love the 324, it's cheap & very useful. Crossover is easily avoided. Don't buy Motorola or bias the output.
>
>But I do wish there were even cheaper opamps with even more corners cut. There must be plenty of uses for opamps with major compromises, eg poor gain. Trouble is things start expensive and drop in price, which would make cruder opamps uncompetitive in the early days. And who wants to sell an even cheaper chip when they can introduce a high ticket one.
>
>
>NT
The Diodes Inc version of the LM324 is 6.5 cents, Digikey price by the
reel. Under 2 cents per opamp.
What does Digikey pay? 5 cents? Why does anybody bother to make and
package a quad opamp for 5 cents?
When these were new, one cost about as much as a couple of bicycles.
https://www.dropbox.com/s/9idlgfug24rqxyb/Philbricks.jpg?raw=1
--
John Larkin Highland Technology, Inc trk
jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com
Reply by ●November 25, 20172017-11-25
On Saturday, November 25, 2017 at 8:48:01 AM UTC+11, John Larkin wrote:
> On Fri, 24 Nov 2017 12:38:20 -0800 (PST),
> bloggs.fredbloggs.fred@gmail.com wrote:
>
> >On Tuesday, November 21, 2017 at 2:35:27 PM UTC-5, Winfield Hill wrote:
> >> John Larkin wrote...
> >> >
> >> > The ancient LM324 has PNP input transistors, and the
> >> > safe input voltages can go to +32 irrespective of V+.
> >>
> >> I also thought of PNP-input single-supply op-amps as
> >> a good solution, but the O.P. seems not to notice.
> >>
> >>
> >> --
> >> Thanks,
> >> - Win
> >
> >Weren't those the original phase inversion cheapos? Meaning you could end up with anything powering them up with a live input.
>
> I don't know of any problems with the original LM324.
>
> Except the ghastly crossover distortion.
>
> And the pitiful slew rate.
>
> And the wimpy drive.
>
> And the horrible stuff that happens if any of the inputs go below
> ground.
>
> The LM709 had front-end zener quirks.
The LM709 had a conventional NPN long-tailed pair as it's input. If you put more than 5V across the inputs one or other base-emitter junction broke down - the voltage is a bit high for a pure Zener breakdown, but reverse-biassing a base-emitter junction to breakdown down damages the base-emitter junction even if it doesn't destroy it.
The data sheet was perfectly explicit about the voltage difference limit between the inputs - it might not have been a feature, but it definitely wasn't a quirk.
Bob Widlar got more ingenious later, and subsequent op amps either used an input stage that could take more volts, or protected the input stage with inverse parallel diodes.
--
Bill Sloman, Sydney
Reply by whit3rd●November 25, 20172017-11-25
On Saturday, November 25, 2017 at 5:28:53 PM UTC-8, tabb...@gmail.com wrote:
> I love the 324, it's cheap & very useful. Crossover is easily avoided. Don't buy Motorola or bias the output.
>
> But I do wish there were even cheaper opamps with even more corners cut.
The ST variant exemplifies why you don't see those; a different op amp has to compete for
the bottom dollar, which is a slow-to-pay-back plan. So, different or not, a low-end
quad op amp gets "LM324" name attached to it, which is how one advertises
the low-end amp while not using an advertising budget.
It has to show up in a quick search for that part number...
Reply by ●November 25, 20172017-11-25
On Sat, 25 Nov 2017 17:28:49 -0800 (PST), tabbypurr@gmail.com wrote:
>On Saturday, 25 November 2017 00:09:27 UTC, John Larkin wrote:
>> On Fri, 24 Nov 2017 14:06:48 -0800 (PST), Leslie Rhorer
>> <rhorerles@gmail.com> wrote:
>> >On Friday, November 24, 2017 at 3:48:01 PM UTC-6, John Larkin wrote:
>> >> I don't know of any problems with the original LM324.
>> >>
>> >> Except the ghastly crossover distortion.
>> >>
>> >> And the pitiful slew rate.
>> >>
>> >> And the wimpy drive.
>> >>
>> >> And the horrible stuff that happens if any of the inputs go below
>> >> ground.
>> >
>> >None of which matter in this situation, which is one reason why your original suggestion is definitely the front runner. I've ordered a couple of Op Amps, including a 324, from Mouser for testing. They should arrive Monday.
>>
>> One other quirk: if one of the four opamps rails, it can mess up the
>> other three. Shared current sources. I don't remember just how bad
>> that is.
>
>I love the 324, it's cheap & very useful. Crossover is easily avoided. Don't buy Motorola or bias the output.
>
>But I do wish there were even cheaper opamps with even more corners cut. There must be plenty of uses for opamps with major compromises, eg poor gain. Trouble is things start expensive and drop in price, which would make cruder opamps uncompetitive in the early days. And who wants to sell an even cheaper chip when they can introduce a high ticket one.
NJR seems to make a business of it. There's always someone who wants
junk. And someone who will buy it.
Reply by ●November 25, 20172017-11-25
On Saturday, 25 November 2017 00:09:27 UTC, John Larkin wrote:
> On Fri, 24 Nov 2017 14:06:48 -0800 (PST), Leslie Rhorer
> <rhorerles@gmail.com> wrote:
> >On Friday, November 24, 2017 at 3:48:01 PM UTC-6, John Larkin wrote:
> >> I don't know of any problems with the original LM324.
> >>
> >> Except the ghastly crossover distortion.
> >>
> >> And the pitiful slew rate.
> >>
> >> And the wimpy drive.
> >>
> >> And the horrible stuff that happens if any of the inputs go below
> >> ground.
> >
> >None of which matter in this situation, which is one reason why your original suggestion is definitely the front runner. I've ordered a couple of Op Amps, including a 324, from Mouser for testing. They should arrive Monday.
>
> One other quirk: if one of the four opamps rails, it can mess up the
> other three. Shared current sources. I don't remember just how bad
> that is.
I love the 324, it's cheap & very useful. Crossover is easily avoided. Don't buy Motorola or bias the output.
But I do wish there were even cheaper opamps with even more corners cut. There must be plenty of uses for opamps with major compromises, eg poor gain. Trouble is things start expensive and drop in price, which would make cruder opamps uncompetitive in the early days. And who wants to sell an even cheaper chip when they can introduce a high ticket one.
NT
Reply by ●November 25, 20172017-11-25
On Friday, 24 November 2017 01:54:53 UTC, David Eather wrote:
> On Fri, 24 Nov 2017 01:54:57 +1000, Joerg <news@analogconsultants.com>
> wrote:
>
> > On 2017-11-22 22:22, Jasen Betts wrote:
> >> On 2017-11-21, Joerg <news@analogconsultants.com> wrote:
> >>> On 2017-11-21 11:35, Winfield Hill wrote:
> >>>> John Larkin wrote...
> >>>>>
> >>>>> The ancient LM324 has PNP input transistors, and the
> >>>>> safe input voltages can go to +32 irrespective of V+.
> >>>>
> >>>
> >>> The datasheet says "Don't do that" though. Looking at the innards on
> >>> page 4 it seems you'd hit a diode path to V+:
> >>>
> >>> http://www.st.com/content/ccc/resource/technical/document/datasheet/bd/fc/46/43/26/8f/40/7f/CD00001046.pdf/files/CD00001046.pdf/jcr:content/translations/en.CD00001046.pdf
> >>
> >> use the TI part.
> >>
> >> http://www.ti.com/lit/ds/snosc16d/snosc16d.pdf
> >>
>
> There is also a lower power version of the lm324 / 358 from ti. It's
> imaginatively called an LP324 and LP358