GAL running hot

classic Classic list List threaded Threaded
29 messages Options
12
Reply | Threaded
Open this post in threaded view
|

GAL running hot

silverdr@wfmh.org.pl
Unfortunately not /that/ GAL.. anyway - I made a small circuit (the first successful one in KiCAD I wrote about not long ago) for the 64 that employs the small 16V8 GAL chip. All works fine but the (Lattice) chip runs surprisingly hot. Does anyone have any experience with GALs and with what can affect their power consumption?

--
SD! (http://e4aws.silverdr.com/)


Reply | Threaded
Open this post in threaded view
|

Re: GAL running hot

Francesco Messineo
On Fri, Apr 20, 2018 at 11:25 AM,  <[hidden email]> wrote:
> Unfortunately not /that/ GAL.. anyway - I made a small circuit (the first successful one in KiCAD I wrote about not long ago) for the 64 that employs the small 16V8 GAL chip. All works fine but the (Lattice) chip runs surprisingly hot. Does anyone have any experience with GALs and with what can affect their power consumption?


In my experience, old PALs and GALs run hot, but I've never really
investigated why. Who would use them nowadays anyway? ;-)

Frank

Reply | Threaded
Open this post in threaded view
|

Re: GAL running hot

Didier Derny
GAL are still easy to find in small quantity  (DIP) and are now inexpensive

it's quite nice for a prototype of a small series

I use them all the time, they give some flexibilty

I also often use max7000s  CPLD PLCC44/PLCC84




Le 20/04/2018 à 14:52, Francesco Messineo a écrit :
> On Fri, Apr 20, 2018 at 11:25 AM,  <[hidden email]> wrote:
>> Unfortunately not /that/ GAL.. anyway - I made a small circuit (the first successful one in KiCAD I wrote about not long ago) for the 64 that employs the small 16V8 GAL chip. All works fine but the (Lattice) chip runs surprisingly hot. Does anyone have any experience with GALs and with what can affect their power consumption?
>
> In my experience, old PALs and GALs run hot, but I've never really
> investigated why. Who would use them nowadays anyway? ;-)
>
> Frank
>


Reply | Threaded
Open this post in threaded view
|

Re: GAL running hot

silverdr@wfmh.org.pl
In reply to this post by Francesco Messineo

> On 2018-04-20, at 14:52, Francesco Messineo <[hidden email]> wrote:
>
>> Unfortunately not /that/ GAL.. anyway - I made a small circuit (the first successful one in KiCAD I wrote about not long ago) for the 64 that employs the small 16V8 GAL chip. All works fine but the (Lattice) chip runs surprisingly hot. Does anyone have any experience with GALs and with what can affect their power consumption?
>
> In my experience, old PALs and GALs run hot,

The datasheet says it should take in the range of 70mA, which makes it less than half W. Yeah, much more than an LS chip but OTOH not /that/ much..

> but I've never really
> investigated why. Who would use them nowadays anyway? ;-)

The fact that e.g. Atmel still have them in their line of products says that I am not the last and only one who would ;-)

Besides - what other options for a small SPLD I have?

--
SD! - http://e4aws.silverdr.com/


Reply | Threaded
Open this post in threaded view
|

Re: GAL running hot

silverdr@wfmh.org.pl
In reply to this post by Didier Derny

> On 2018-04-20, at 17:10, didier derny <[hidden email]> wrote:
>
> GAL are still easy to find in small quantity  (DIP) and are now inexpensive

And PLCC too. Atmel still has them.

> it's quite nice for a prototype of a small series
>
> I use them all the time, they give some flexibilty
>
> I also often use max7000s  CPLD PLCC44/PLCC84

True, but for the purposes here a CPLD with even 44 pins only is already an overkill. The one I use has 20 and a few are still left unused. OTOH if I wanted to build the same logic with 74 series, it'd eat a lot of board's real estate.

--
SD! - http://e4aws.silverdr.com/


Reply | Threaded
Open this post in threaded view
|

Re: GAL running hot

MiaM
In reply to this post by Didier Derny
Den Fri, 20 Apr 2018 17:10:57 +0200 skrev didier derny
<[hidden email]>:
> GAL are still easy to find in small quantity  (DIP) and are now
> inexpensive
>
> it's quite nice for a prototype of a small series
>
> I use them all the time, they give some flexibilty
>
> I also often use max7000s  CPLD PLCC44/PLCC84

I think a small production run of adapter boards for using those
hard-to-solder-fine-pitch CPLD's with DIL sockets (or similar) might be
something worth investigating?

--
(\_/) Copy the bunny to your mails to help
(O.o) him achieve world domination.
(> <) Come join the dark side.
/_|_\ We have cookies.

Reply | Threaded
Open this post in threaded view
|

Re: GAL running hot

Gerrit Heitsch
In reply to this post by silverdr@wfmh.org.pl
On 04/20/2018 05:56 PM, [hidden email] wrote:
>
>> On 2018-04-20, at 14:52, Francesco Messineo <[hidden email]> wrote:
>>
>>> Unfortunately not /that/ GAL.. anyway - I made a small circuit (the first successful one in KiCAD I wrote about not long ago) for the 64 that employs the small 16V8 GAL chip. All works fine but the (Lattice) chip runs surprisingly hot. Does anyone have any experience with GALs and with what can affect their power consumption?
>>
>> In my experience, old PALs and GALs run hot,
>
> The datasheet says it should take in the range of 70mA, which makes it less than half W. Yeah, much more than an LS chip but OTOH not /that/ much..

Yes, but half a watt in such a small case gets quite hot. A 6569 VIC in
a C64 uses about 1.5W (200mA/5V and about 40mA/12V) and gets pretty hot
(above 60C on the surface without heatsink).

  Gerrit


Reply | Threaded
Open this post in threaded view
|

Re: GAL running hot

silverdr@wfmh.org.pl

> On 2018-04-20, at 18:41, Gerrit Heitsch <[hidden email]> wrote:
>
>>>> Unfortunately not /that/ GAL.. anyway - I made a small circuit (the first successful one in KiCAD I wrote about not long ago) for the 64 that employs the small 16V8 GAL chip. All works fine but the (Lattice) chip runs surprisingly hot. Does anyone have any experience with GALs and with what can affect their power consumption?
>>>
>>> In my experience, old PALs and GALs run hot,
>> The datasheet says it should take in the range of 70mA, which makes it less than half W. Yeah, much more than an LS chip but OTOH not /that/ much..
>
> Yes, but half a watt in such a small case gets quite hot. A 6569 VIC in a C64 uses about 1.5W (200mA/5V and about 40mA/12V) and gets pretty hot (above 60C on the surface without heatsink).

True, and the case is small - especially that it's a PLCC - so it might actually be something to learn to live with. I am still wondering if the power usage could be noticeably depending on the way it is programmed, which pins are used, etc. I'll check the Atmel once I get a programmer that can handle them. I needed to use Lattice as it was the only type my programmer could work reliably with but Atmel advertises lower currents in their datasheet.

--
SD! - http://e4aws.silverdr.com/


Reply | Threaded
Open this post in threaded view
|

Re: GAL running hot

Gerrit Heitsch
On 04/20/2018 07:00 PM, [hidden email] wrote:

>
>> On 2018-04-20, at 18:41, Gerrit Heitsch <[hidden email]> wrote:
>>
>>>>> Unfortunately not /that/ GAL.. anyway - I made a small circuit (the first successful one in KiCAD I wrote about not long ago) for the 64 that employs the small 16V8 GAL chip. All works fine but the (Lattice) chip runs surprisingly hot. Does anyone have any experience with GALs and with what can affect their power consumption?
>>>>
>>>> In my experience, old PALs and GALs run hot,
>>> The datasheet says it should take in the range of 70mA, which makes it less than half W. Yeah, much more than an LS chip but OTOH not /that/ much..
>>
>> Yes, but half a watt in such a small case gets quite hot. A 6569 VIC in a C64 uses about 1.5W (200mA/5V and about 40mA/12V) and gets pretty hot (above 60C on the surface without heatsink).
>
> True, and the case is small - especially that it's a PLCC - so it might actually be something to learn to live with. I am still wondering if the power usage could be noticeably depending on the way it is programmed, which pins are used, etc. I'll check the Atmel once I get a programmer that can handle them. I needed to use Lattice as it was the only type my programmer could work reliably with but Atmel advertises lower currents in their datasheet.

Well, if you screw up, you might cause 2 output drivers work against
each other. That would increase power consumption by quite a bit.
Otherwise it also depends on the load on the output drivers. Do they
drive only MOS-inputs or are there pullups involved?

  Gerrit


Reply | Threaded
Open this post in threaded view
|

Re: GAL running hot

Francesco Messineo
In reply to this post by Didier Derny
On Fri, Apr 20, 2018 at 5:10 PM, didier derny <[hidden email]> wrote:
> GAL are still easy to find in small quantity  (DIP) and are now inexpensive
>
> it's quite nice for a prototype of a small series
>
> I use them all the time, they give some flexibilty
>
> I also often use max7000s  CPLD PLCC44/PLCC84

I concur, I like very much the max7000 series but also ATF150x from
Atmel are very good imho.
Yes, bigger than a typical PAL/GAL indeed.

Frank

Reply | Threaded
Open this post in threaded view
|

Re: GAL running hot

silverdr@wfmh.org.pl
In reply to this post by Gerrit Heitsch

> On 2018-04-20, at 19:30, Gerrit Heitsch <[hidden email]> wrote:
>
>>>>>> Unfortunately not /that/ GAL.. anyway - I made a small circuit (the first successful one in KiCAD I wrote about not long ago) for the 64 that employs the small 16V8 GAL chip. All works fine but the (Lattice) chip runs surprisingly hot. Does anyone have any experience with GALs and with what can affect their power consumption?
>>>>>
>>>>> In my experience, old PALs and GALs run hot,
>>>> The datasheet says it should take in the range of 70mA, which makes it less than half W. Yeah, much more than an LS chip but OTOH not /that/ much..
>>>
>>> Yes, but half a watt in such a small case gets quite hot. A 6569 VIC in a C64 uses about 1.5W (200mA/5V and about 40mA/12V) and gets pretty hot (above 60C on the surface without heatsink).
>> True, and the case is small - especially that it's a PLCC - so it might actually be something to learn to live with. I am still wondering if the power usage could be noticeably depending on the way it is programmed, which pins are used, etc. I'll check the Atmel once I get a programmer that can handle them. I needed to use Lattice as it was the only type my programmer could work reliably with but Atmel advertises lower currents in their datasheet.
>
> Well, if you screw up, you might cause 2 output drivers work against each other. That would increase power consumption by quite a bit. Otherwise it also depends on the load on the output drivers. Do they drive only MOS-inputs or are there pullups involved?

They drive only CMOS inputs of an E(E)PROM. I left the unused pins undefined and unconnected. The way I understood the docs that should be fine but maybe defining/connecting them change something.

--
SD! - http://e4aws.silverdr.com/


Reply | Threaded
Open this post in threaded view
|

Re: programming Atmel GALs (was: GAL running hot)

silverdr@wfmh.org.pl
In reply to this post by silverdr@wfmh.org.pl


> On 2018-04-20, at 19:00, [hidden email] wrote:
>
> I'll check the Atmel once I get a programmer that can handle them. I needed to use Lattice as it was the only type my programmer could work reliably with but Atmel advertises lower currents in their datasheet.

Theoretically I got one (the G540) but it fails very similarly to the previous one. Meaning I still can't program Atmel chips. Anyone knows a proven to work method/device for programming those? Except GALBlast, which is also said not to be trusted... ?

--
SD! - http://e4aws.silverdr.com/


Reply | Threaded
Open this post in threaded view
|

Re: programming Atmel GALs (was: GAL running hot)

Rainer Buchty
On Fri, 25 May 2018, [hidden email] wrote:

> Theoretically I got one (the G540) but it fails very similarly to the
> previous one. Meaning I still can't program Atmel chips. Anyone knows
> a proven to work method/device for programming those? Except GALBlast,
> which is also said not to be trusted... ?

The ATFs have different programming algorithms. Seems like this
information

  http://www.ecstaticlyrics.com/electronics/ATF22V10C/

has now spread around a bit and was merged into "ATFblast" which is an
alternative software to the GALblast hardware.

  http://www.bhabbott.net.nz/atfblast.html

This one has also a nice writeup of the differences to "standard" GALs.

Makes me wonder why it hasn't found its way into the TL866 software yet,
which supported the ATF16V8 (but not the ATF22V10) for years.

Rainer


Reply | Threaded
Open this post in threaded view
|

Re: programming Atmel GALs (was: GAL running hot)

silverdr@wfmh.org.pl


> On 2018-05-25, at 17:06, Rainer Buchty <[hidden email]> wrote:
>
> Makes me wonder why it hasn't found its way into the TL866 software yet, which supported the ATF16V8 (but not the ATF22V10) for years.

TL866 programs Lattice 16V8 chips but fails on Atmel's ones here. G540 does the same. I am now starting to wonder whether I haven't got a whole batch of faulty chips... I tried a few, randomly chosen, but they all behaved the same. OTOH the Chinese software for those doesn't really inspire confidence in its reliability.

--
SD! - http://e4aws.silverdr.com/


Reply | Threaded
Open this post in threaded view
|

Re: programming Atmel GALs

Michał Pleban
Hello!

[hidden email] wrote:

> TL866 programs Lattice 16V8 chips but fails on Atmel's ones here. G540 does the same. I am now starting to wonder whether I haven't got a whole batch of faulty chips... I tried a few, randomly chosen, but they all behaved the same. OTOH the Chinese software for those doesn't really inspire confidence in its reliability.

Over a long time, I tried many different programmers and they all could
program Lattice GAL's but not Atmel GALs. There must be something
inherently different with these Atmel chips.

Regards,
Michau


Reply | Threaded
Open this post in threaded view
|

Re: programming Atmel GALs

Ruud
Hallo Patryk,


> TL866 programs Lattice 16V8 chips but fails on Atmel's ones here.

I have a LabProg+ from Elnec. According the menu it supports the
ATF16V8B and other variants although I have no adapters to program
the "square" version. You could send me one or two and I try if I
can program them for you.


--
   
Kind regards / Met vriendelijke groet, Ruud Baltissen
www.Baltissen.org



Reply | Threaded
Open this post in threaded view
|

Re: programming Atmel GALs (was: GAL running hot)

dave_m
In reply to this post by silverdr@wfmh.org.pl
>
Meaning I still can't program Atmel chips. Anyone knows a proven to work
method/device for programming those?
>

You should use the Atmel version of WinCupl to generate your PAL Design
Specification file and output the JEDEC file. To program the 16V8 and 22V10
they used to recommend the Wellon VP-290 which cost about $160 USD. The
latest version may be the VP-299. The combination of the Atmel Wincupl and
Wellon works well.

Below is the link to the free Atmel WinCupl software.

Microchip (ATMEL)
<http://www.microchip.com/design-centers/programmable-logic/spld-cpld/tools/software/wincupl>  

-Dave



--
Sent from: http://cbm-hackers.2304266.n4.nabble.com/

Reply | Threaded
Open this post in threaded view
|

Re: programming Atmel GALs

Michał Pleban
Hello!

dave_m wrote:

> You should use the Atmel version of WinCupl to generate your PAL Design
> Specification file and output the JEDEC file. To program the 16V8 and 22V10
> they used to recommend the Wellon VP-290 which cost about $160 USD. The
> latest version may be the VP-299. The combination of the Atmel Wincupl and
> Wellon works well.

No, it's not the problem of JEDEC file. I created it with Atmel WinCUPL,
but the programmer cannot prpgram Atmel chips with it (even though it
can program Lattice chips with this JED file).

Regards,
Michau.


Reply | Threaded
Open this post in threaded view
|

Re: programming Atmel GALs

silverdr@wfmh.org.pl
In reply to this post by Michał Pleban

> On 2018-05-26, at 13:06, Michał Pleban <[hidden email]> wrote:
>
>> TL866 programs Lattice 16V8 chips but fails on Atmel's ones here. G540 does the same. I am now starting to wonder whether I haven't got a whole batch of faulty chips... I tried a few, randomly chosen, but they all behaved the same. OTOH the Chinese software for those doesn't really inspire confidence in its reliability.
>
> Over a long time, I tried many different programmers and they all could
> program Lattice GAL's but not Atmel GALs. There must be something
> inherently different with these Atmel chips.

They definitely are different and need to be programmed differently. There are no doubts about it. The doubts and problem are that programmers/software, which are supposed to support programming Atmel chips fail miserably here. Do I understand correctly that you also used programmers, which - according to the specs - should be able to program Atmel GALs but they didn't?

--
SD! - http://e4aws.silverdr.com/


Reply | Threaded
Open this post in threaded view
|

Re: programming Atmel GALs

silverdr@wfmh.org.pl
In reply to this post by Ruud

> On 2018-05-26, at 14:48, [hidden email] wrote:
>
>> TL866 programs Lattice 16V8 chips but fails on Atmel's ones here.
>
> I have a LabProg+ from Elnec. According the menu it supports the
> ATF16V8B and other variants although I have no adapters to program
> the "square" version. You could send me one or two and I try if I
> can program them for you.

Ruud, thank you - I have now one more - an order of magnitude more expensive one (arrghh) - programmer on its way. If that one also fails then I shall get back to you with the chips to check on your programmer.

--
SD! - http://e4aws.silverdr.com/


12