Reply by Phil Hobbs ●December 22, 20212021-12-22
g c wrote:

> On Wednesday, 22 December 2021 at 1:51:19 pm UTC+11, Ed Lee wrote:
>> On Tuesday, December 21, 2021 at 6:34:26 PM UTC-8, g c wrote:
>>> On Wednesday, 22 December 2021 at 10:38:53 am UTC+11, Ed Lee wrote:
>>>> On Tuesday, December 21, 2021 at 3:29:22 PM UTC-8, g c wrote:
>>>>> Hi
>>>>>
>>>>> LM2676S-ADJ lead-free are hard to get now. Has anyone tried substituting LM2670 series with success? I am looking at the datasheets trying to find a trap. Not using sync. Need a direct substitute without any changes to surrounding passive parts if possible.
>>>> Fixed voltage version seems to be in better supply. What voltage is your design?
>>> 16V
>> You can use the 12V version and use around 7K external R2. No R1 and no change to your circuit board.
> 
> I'll look at that option, thanks.
> 


Watch out for thermal drift and initial accuracy, though--if the on-chip 
resistors are diffusions or polysilicon rather than metal, they're 
typically +-30% initially and have tempcos between 500 and about 5000 
ppm/K.  (See e.g. 
<https://pallen.ece.gatech.edu/Academic/ECE_4430/Summer_2004/L194-CMOSPassCompII(2-UP).pdf>.)

You'd probably be better off keeping the voltage divider but reducing 
its impedance level by 10x or so (and changing the ratio, of course). 
That way you're basically taking ratios of resistors of the same type.

Before there were LM317s, people used to do that with LM309s and 7805s.

Cheers

Phil Hobbs

-- 
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com
Reply by g c ●December 22, 20212021-12-22
On Wednesday, 22 December 2021 at 1:51:19 pm UTC+11, Ed Lee wrote:

> On Tuesday, December 21, 2021 at 6:34:26 PM UTC-8, g c wrote: 
> > On Wednesday, 22 December 2021 at 10:38:53 am UTC+11, Ed Lee wrote: 
> > > On Tuesday, December 21, 2021 at 3:29:22 PM UTC-8, g c wrote: 
> > > > Hi 
> > > > 
> > > > LM2676S-ADJ lead-free are hard to get now. Has anyone tried substituting LM2670 series with success? I am looking at the datasheets trying to find a trap. Not using sync. Need a direct substitute without any changes to surrounding passive parts if possible. 
> > > Fixed voltage version seems to be in better supply. What voltage is your design? 
> > 16V
> You can use the 12V version and use around 7K external R2. No R1 and no change to your circuit board.


I'll look at that option, thanks.
Reply by Ed Lee ●December 21, 20212021-12-21
On Tuesday, December 21, 2021 at 6:34:26 PM UTC-8, g c wrote:

> On Wednesday, 22 December 2021 at 10:38:53 am UTC+11, Ed Lee wrote: 
> > On Tuesday, December 21, 2021 at 3:29:22 PM UTC-8, g c wrote: 
> > > Hi 
> > > 
> > > LM2676S-ADJ lead-free are hard to get now. Has anyone tried substituting LM2670 series with success? I am looking at the datasheets trying to find a trap. Not using sync. Need a direct substitute without any changes to surrounding passive parts if possible. 
> > Fixed voltage version seems to be in better supply. What voltage is your design?
> 16V


You can use the 12V version and use around 7K external R2.  No R1 and no change to your circuit board.
Reply by g c ●December 21, 20212021-12-21
Hi

LM2676S-ADJ  lead-free are hard to get now. Has anyone tried substituting LM2670 series with success? I am looking at the datasheets trying to find a trap. Not using sync. Need a direct substitute without any changes to surrounding passive parts if possible.

Geoff