Alistair George wrote:
> Jim as an aside, I know you have your own hardware emulators, but have 
> you used the DS2250? I built a bunch of these emulators up 8 years ago, 
> when they worked perfectly, now cant get them to work, despite various 
> RS232 cables, baud rates, DOS (not XP) and so on - keep getting coms 
> errors, but can use dallas KIT5k to upload kernal.
> Al.
Fixed it was a pot puri of various things from a defective coms cable 
(which I'd checked) Windows sometimes doing coms reliably, and other 
times not.
Whew that was tricky!
Al.

On Sun, 25 Jun 2006 10:52:12 +0100, the renowned Ian Bell
<ruffrecords@yahoo.com> wrote:

>Jim Granville wrote:
>
>> Alistair George wrote:
>> 
>>> Ian Bell wrote:
>>> 
>>>> Alistair George wrote:
>>>>
>>>>> Re HEF40106BT been there, tried that does not work in our use, which as
>>>>> a very high gain oscillator.
>>>>>
>>>>
>>>> You are using a hex schmitt trigger as a high gain oscillator?
>>>>
>>>> Ian
>>> 
>>> Yes, in the original National app notes shows a schmitt trigger as an
>>> oscillator. Variations between manufacturers have largely made that form
>>> of use redundant.
>> 
>>   This is curious - I take it you mean something other than the
>> standard, widely used  2 terminal RC Schmitt oscillator ( or the 3
>> terminal oscillator, which is more acurate but needs 3 gates of a 40106 )
>> ?
>> 
>
>I think he means in a crystal oscillator - ISTR this was a common way of
>generating a microprocessor clock in the 80s but even then it was known to
>be unreliable due to using a digital device in an analog mode.
>
>Ian

IIRC, the 8048/8749 etc. NMOS microcontrollers used a ST in a Pierce
oscillator configuration. IIRC it was supposed to start up in RC mode
@ 50kHz or so and then transition to oscillating at the crystal
frequency. But there are potential problems (it might not make that
transition depending on parameters) and that configuration isn't seen
much these days. 


Best regards, 
Spehro Pefhany
-- 
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Jim Granville wrote:

> Alistair George wrote:
> 
>> Ian Bell wrote:
>> 
>>> Alistair George wrote:
>>>
>>>> Re HEF40106BT been there, tried that does not work in our use, which as
>>>> a very high gain oscillator.
>>>>
>>>
>>> You are using a hex schmitt trigger as a high gain oscillator?
>>>
>>> Ian
>> 
>> Yes, in the original National app notes shows a schmitt trigger as an
>> oscillator. Variations between manufacturers have largely made that form
>> of use redundant.
> 
>   This is curious - I take it you mean something other than the
> standard, widely used  2 terminal RC Schmitt oscillator ( or the 3
> terminal oscillator, which is more acurate but needs 3 gates of a 40106 )
> ?
> 

I think he means in a crystal oscillator - ISTR this was a common way of
generating a microprocessor clock in the 80s but even then it was known to
be unreliable due to using a digital device in an analog mode.

Ian

Jim as an aside, I know you have your own hardware emulators, but have 
you used the DS2250? I built a bunch of these emulators up 8 years ago, 
when they worked perfectly, now cant get them to work, despite various 
RS232 cables, baud rates, DOS (not XP) and so on - keep getting coms 
errors, but can use dallas KIT5k to upload kernal.
Al.

Jim Granville wrote:
> Alistair George wrote:
> 
>> Ian Bell wrote:
>>
>>> Alistair George wrote:
>>>
>>>> Re HEF40106BT been there, tried that does not work in our use, which as
>>>> a very high gain oscillator.
>>>>
>>>
>>> You are using a hex schmitt trigger as a high gain oscillator?
>>>
>>> Ian
>>
>> Yes, in the original National app notes shows a schmitt trigger as an 
>> oscillator. Variations between manufacturers have largely made that 
>> form of use redundant.
> 
>  This is curious - I take it you mean something other than the
> standard, widely used  2 terminal RC Schmitt oscillator ( or the 3 
> terminal oscillator, which is more acurate but needs 3 gates of a 40106 ) ?
> 
>  Do you have a URL of such a configuration ?
> 
> -jg
> 
> 
No Jim, the former as you suggest but in our app it uses a 3m3 feedback 
r and a special capacitor. Spreads between i/c of the same maker were 
acceptable, but not between supposed replacement alternatives.
In hindsight, I should have used a comparator for the job.
Al.

Alistair George wrote:

> Ian Bell wrote:
> 
>> Alistair George wrote:
>>
>>> Re HEF40106BT been there, tried that does not work in our use, which as
>>> a very high gain oscillator.
>>>
>>
>> You are using a hex schmitt trigger as a high gain oscillator?
>>
>> Ian
> 
> Yes, in the original National app notes shows a schmitt trigger as an 
> oscillator. Variations between manufacturers have largely made that form 
> of use redundant.

  This is curious - I take it you mean something other than the
standard, widely used  2 terminal RC Schmitt oscillator ( or the 3 
terminal oscillator, which is more acurate but needs 3 gates of a 40106 ) ?

  Do you have a URL of such a configuration ?

-jg

Ian Bell wrote:
> Alistair George wrote:
> 
>> Re HEF40106BT been there, tried that does not work in our use, which as
>> a very high gain oscillator.
>>
> 
> You are using a hex schmitt trigger as a high gain oscillator?
> 
> Ian
Yes, in the original National app notes shows a schmitt trigger as an 
oscillator. Variations between manufacturers have largely made that form 
of use redundant.


PS thanks to others for continuing this thread to the enlightenment of 
myself and others on the use of, and alternatives to the original 8051 
series of uP.
Al.

Chris Hills wrote:
<snip>  I
> think it is the most popular MCU on the planet including x86. AFAIK most
> PC's have one or two  51's in them somewhere. Used to be keyboard and
> mouse. 

  Yes, SST have an interesting new device, targets PCs, that has the
4MBit BOOT FLASH, and the FLASH C51 IO controller, all for $3.50@10K

[also usefull for Loggers, or just lots of IO ]

http://www.sst.com/news/?id=320

another notable announcement is ASIX's 10/100 Ethernet, with inbuilt PHY
and TCP/IP, and a 100MHz C51 core, and up to 512KF/32KR.
I have not seen price indicators on this one yet.

http://www.asix.com.tw/products.php?op=pItemdetail&PItemID=91;72;103

-jg

In article <449d6daa$0$15791$14726298@news.sunsite.dk>, Bubb
<bubb@telia.se> writes
>What about a soft core 8051 written in VHDL or Verilog? Reusable in an FPGA
>for a long time to come. :-)

Lets face it the 8051 will carry on for a long while yet.  There are
VERY many producers of 8051 parts and cores, never mind a second source
you have multiple. The 8051 core is low cost if you want to put it in an
FPGA to create your own.

There are many high quality tools and a lot of experience for the 8051
out there.  The packages range from 8 pin SMT to 150+ pins though I
think the 40 pin DILs are hard to find now...  Even radiation hardened
versions. 

I am sure that the 600 odd variants will shrink down to only a few
hundred over time but I can't see it dying out for a few years yet.  I
think it is the most popular MCU on the planet including x86. AFAIK most
PC's have one or two  51's in them somewhere. Used to be keyboard and
mouse. 
 

>"Alistair George" <noname@xtra.co.nz> skrev i meddelandet
>news:449c5480@news.orcon.net.nz...
>> Hi All.
>> I used Atmel 8051 derivatives in all my earlier projects. Atmel made a
>> really nice chip in the AT89C55 then later they changed the die and it
>> was not as good; chip 'engineers' come along and change a tried and true
>> chip and its specs change and then it does not work properly anymore. EG
>> CD40106/74C14 performs totally differently than the original National
>> Semiconductors CD40106.
>> But thats not the subject of this post ;-)
>>
>> .....................
>>
>> I have noticed that in my latest suppliers linecard that only the
>> AT89c2051 is being sold and it seems the 'old' 8051 is a dying breed.
>> Even to source the AT89C55WD is a special order here in New Zealand.
>> Does this mean the 8051/2 series is considered defunct? Should I be
>> looking at alternatives for my next product? If so, its a pain as I have
>> programmer, emulator and software to achieve my goals with the 8051
>series.
>> Thank you.
>> Alistair.
>
>

-- 
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills  Staffs  England     /\/\/\/\/
/\/\/ chris@phaedsys.org      www.phaedsys.org \/\/\
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

What about a soft core 8051 written in VHDL or Verilog? Reusable in an FPGA
for a long time to come. :-)

"Alistair George" <noname@xtra.co.nz> skrev i meddelandet
news:449c5480@news.orcon.net.nz...
> Hi All.
> I used Atmel 8051 derivatives in all my earlier projects. Atmel made a
> really nice chip in the AT89C55 then later they changed the die and it
> was not as good; chip 'engineers' come along and change a tried and true
> chip and its specs change and then it does not work properly anymore. EG
> CD40106/74C14 performs totally differently than the original National
> Semiconductors CD40106.
> But thats not the subject of this post ;-)
>
> .....................
>
> I have noticed that in my latest suppliers linecard that only the
> AT89c2051 is being sold and it seems the 'old' 8051 is a dying breed.
> Even to source the AT89C55WD is a special order here in New Zealand.
> Does this mean the 8051/2 series is considered defunct? Should I be
> looking at alternatives for my next product? If so, its a pain as I have
> programmer, emulator and software to achieve my goals with the 8051
series.
> Thank you.
> Alistair.