Changing refresh rate for DRAM while in operation?

[sendthis]

Hi,

I'm trying to control a SDR SDRAM (Micron 64Mbit chip) using an Altera
DE2 board. I've gotten the hardware interface squared away (thanks
everyone for your help!).

Now it's the tricky stuff. Any one have an idea how I can change the
refresh rate while the RAM is in operation?

I have the DRAM interface built using the SOPC builder that comes with
Quartus II using the NIOS II system.

I know you can change the refresh rate during the build but I need a
way to change the refresh rate during operation. The only thing I can
think of is maybe change the clock speed? I have it running off a
50Mhz clock....

Thanks,
Eric

 

[KJ]

Hi,

I'm trying to control a SDR SDRAM (Micron 64Mbit chip) using an Altera
DE2 board. I've gotten the hardware interface squared away (thanks
everyone for your help!).

Now it's the tricky stuff. Any one have an idea how I can change the
refresh rate while the RAM is in operation?

The most obvious question would be 'Why?'

I have the DRAM interface built using the SOPC builder that comes with
Quartus II using the NIOS II system.

That will limit your options (as would probably most other vendor IP DRAM
controllers).

I know you can change the refresh rate during the build but I need a
way to change the refresh rate during operation. The only thing I can
think of is maybe change the clock speed? I have it running off a
50Mhz clock....

A simpler way would be to simply have a DRAM controller that has an explicit
'Refresh Request' input that would cause the controller to perform a
refresh. Then connect that input up to any programmable timer or other
logic that you would like to use. Changing the clock rate would be far down
on my list of ways to accomplish your goal....but again, it begs the
original question about why you would want to change the refresh rate
dynamically at all.

KJ

 

[David Spencer]

Now it's the tricky stuff. Any one have an idea how I can change the

refresh rate while the RAM is in operation?

Why?

 

[Jim Stewart]

The most obvious question would be 'Why?'

That will limit your options (as would probably most other vendor IP DRAM
controllers).

A simpler way would be to simply have a DRAM controller that has an explicit
'Refresh Request' input that would cause the controller to perform a
refresh. Then connect that input up to any programmable timer or other
logic that you would like to use. Changing the clock rate would be far down
on my list of ways to accomplish your goal....but again, it begs the
original question about why you would want to change the refresh rate
dynamically at all.

Assuming he has a good reason to change it,
the safest thing to do would be to call a
routine in flash to change it.

 

[CBFalconer]

I'm trying to control a SDR SDRAM (Micron 64Mbit chip) using an
Altera DE2 board. I've gotten the hardware interface squared away
(thanks everyone for your help!).

Now it's the tricky stuff. Any one have an idea how I can change
the refresh rate while the RAM is in operation?

I have the DRAM interface built using the SOPC builder that comes
with Quartus II using the NIOS II system.

I know you can change the refresh rate during the build but I need
a way to change the refresh rate during operation. The only thing
I can think of is maybe change the clock speed? I have it running
off a 50Mhz clock....

Since the only purpose of the refresh circuitry is to avoid the
memory dropping bits, it should already be running at the slowest
possible rate, and speed reduction will be harmful, while speed
increase will do no good. So this is not a good idea.

What are you trying to do?

 

[MitchAlsup]

Since the only purpose of the refresh circuitry is to avoid the
memory dropping bits, it should already be running at the slowest
possible rate, and speed reduction will be harmful, while speed
increase will do no good. So this is not a good idea.

I disagree (softly), having designed several memory controllers, I
always found it easier to just insert a READ DATA command into the
DRAM when a refresh was needed, rather than insert a refresh command.
The timing differences between refresh and a loosly coupled string of
READS is such that one can refresh ahead with READs easier and then be
in a position to absorb a longer string of demand requests by not
using the REFRESH commands. Thus while running at the slowest overall
rate, one can bunch and distribute the refresh mechanics to better
interleave same with the demand memory requests and gain something.

But I will state the overall performance differences are a fraction of
the refresh overhead anyways.

What are you trying to do?

That is the real question.

 

[Paul Keinanen]

But I will state the overall performance differences are a fraction of
the refresh overhead anyways.


That is the real question.

If the idea is to reduce the refresh overhead on a busy bus, reducing
the relatively slow refresh rate does not make much sense.

However, if the memory content is to be maintained for a long time
without any data access in a battery powered device, it would make
sense to reduce the refresh rate at low ambient temperatures. The high
refresh rates are needed at the top end of the temperature range, but
at lower temperatures, a slower refresh rate would be sufficient,
which reduces the power consumption and increase battery life.
Unfortunately, refresh rate figures are seldom available for lower
temperatures.

Paul

 

[Sean Durkin]

However, if the memory content is to be maintained for a long time
without any data access in a battery powered device, it would make
sense to reduce the refresh rate at low ambient temperatures. The high
refresh rates are needed at the top end of the temperature range, but
at lower temperatures, a slower refresh rate would be sufficient,
which reduces the power consumption and increase battery life.
Unfortunately, refresh rate figures are seldom available for lower
temperatures.

If you were really aiming for long run time on battery power, I suppose
you'd just use DRAM devices specifically made for such an application.

Mobile SDRAM devices often have a temperature compensated self refresh
feature. You just enter a special suspend mode and the device does the
refresh itself, and only as often as required according to the current
temperature. You can also tell it to just refresh a part of the memory
array, in case you don't use it all.

This is usually way better than anything one could do on his/her own.

So, the question still stands: What does the OP really want to do?

cu,
Sean

 

[CBFalconer]

I disagree (softly), having designed several memory controllers,
I always found it easier to just insert a READ DATA command into
the DRAM when a refresh was needed, rather than insert a refresh
command. The timing differences between refresh and a loosly
coupled string of READS is such that one can refresh ahead with
READs easier and then be in a position to absorb a longer string
of demand requests by not using the REFRESH commands. Thus while
running at the slowest overall rate, one can bunch and distribute
the refresh mechanics to better interleave same with the demand
memory requests and gain something.

But I will state the overall performance differences are a
fraction of the refresh overhead anyways.


That is the real question.

Since the OP seems to have disappeared to wherever OPs go, I
suspect we will never find out.

 

[David Spencer]

Since the OP seems to have disappeared to wherever OPs go, I
suspect we will never find out.

Don't you just hate it when that happens? Even if the OP now realises that
what he was trying to do wasn't appropriate or necessary, it would be nice
if he just explained his original intentions to us.

 

[Gabor]

Don't you just hate it when that happens? Even if the OP now realises that
what he was trying to do wasn't appropriate or necessary, it would be nice
if he just explained his original intentions to us.

He's probably sorry about the flame war he unintentionally created
but thinks it would have been nice if one of the 25 replies answered
his original question...

 

[KJ]

On Oct 25, 12:59 pm, "David Spencer" <[email protected]>
but thinks it would have been nice if one of the 25 replies answered
his original question...

The very first reply to the original post did answer the OP's original
question.

The rest of the thread has been for entertainment and educational
value.

KJ

 

[sendthis]

Since the OP seems to have disappeared to wherever OPs go, I
suspect we will never find out.

I didn't disappear, I posted a reply but for some reason it didn't
show up... I didn't want to accidentally spam the newsgroups by
reposting and figured I'd wait to make sure it wasn't just my
newsreader or ISP causing the problem.

Anyway, I guess I'll answer the reason why I want to do this in the
same post.

I'm trying to characterize a DRAM device in certain environmental
(radiation) conditions and see how that effects the retention
characteristics. I'm not sure if there tests the industry uses to do
this, but I needed to evaluate it realtime.

I'm using the core Altera provided but all the code is there (except
for the NIOS II cpu). So I have direct access to the SDRAM
controller.

 

[sendthis]

Assuming he has a good reason to change it,
the safest thing to do would be to call a
routine in flash to change it.- Hide quoted text -

Thanks for giving me the benefit of the doubt. I put this in another
reply, but the reason I want to change it is to characterize the DRAM
to get it's retention characteristics in a radiation environment. So I
want to know how long the DRAM keeps it's charge given a specific and
controlled environment.

Interfacing, programming, and DRAM definitely aren't my areas of study
(materials is). This is for a small but time consuming part of my
thesis project.

Thanks,
eric

 

[sendthis]

Probably so, but it isn't at all obvious how to answer. The DRAM
doesn't care as long as every row is refreshed within the specified
amount of time. Some refresh all rows in a big burst, others one
at a time uniformly over the interval. You can refresh faster than
the specified rate, but there is no system independent way to
describe how to do that. For systems with a variable speed
clock (such as power saving modes) one does have to design
the refresh system appropriately.

I know the mode register is initialized at the beginning with the
refresh rate (and some other information). Is it possible to reload
the mode register and will this do anything to the stored data (such
as letting all the caps discharge)? Is this even possible?

I do appreciate everyone's replies and I certainly didn't mean to
ignore your answers and questions that were trying to help me.


Paul mentioned in his reply that it makes sense to do it in different
temperatures. This really is similar to what I am trying to do. I'm
trying to figure out (partly) if the refresh rate will help with the
radiation tolerance of the device (i.e. speeding it up).

 

[glen herrmannsfeldt]

Gabor wrote:

(snip)

He's probably sorry about the flame war he unintentionally created
but thinks it would have been nice if one of the 25 replies answered
his original question...

Probably so, but it isn't at all obvious how to answer. The DRAM
doesn't care as long as every row is refreshed within the specified
amount of time. Some refresh all rows in a big burst, others one
at a time uniformly over the interval. You can refresh faster than
the specified rate, but there is no system independent way to
describe how to do that. For systems with a variable speed
clock (such as power saving modes) one does have to design
the refresh system appropriately.

-- glen

 

[Peter Alfke]

I didn't disappear, I posted a reply but for some reason it didn't
show up... I didn't want to accidentally spam the newsgroups by
reposting and figured I'd wait to make sure it wasn't just my

Here is a free suggestion (the price is right):
I would write a specific word-pattern with an even mix of 1 and 0 into
every location in the whole DRAM.
Then read back sequentially at a slow pace through all addresses,
always checking the readback.
Sooner or later, you will pick up an error, becaue you exceeded the
refresh delay.
You may want to repeat this with different starting addresses and with
different word patterns.

My opinion:
SEU errors have little to do with refresh delay, since an ion can tip
over even a fully charged bit. But that seems to be what you want to
find out...
Peter Alfke

 

[Antti]

On Oct 25, 4:20 pm, (e-mail address removed) wrote:> > Since the OP seems to have disappeared to wherever OPs go, I


Here is a free suggestion (the price is right):
I would write a specific word-pattern with an even mix of 1 and 0 into
every location in the whole DRAM.
Then read back sequentially at a slow pace through all addresses,
always checking the readback.
Sooner or later, you will pick up an error, becaue you exceeded the
refresh delay.
You may want to repeat this with different starting addresses and with
different word patterns.

My opinion:
SEU errors have little to do with refresh delay, since an ion can tip
over even a fully charged bit. But that seems to be what you want to
find out...
Peter Alfke







- Zitierten Text anzeigen -

eh, there used to be a special software that allows a DRAM to work as
black white camera

it was using 64K x 1 military DRAMs, with REMOVED top lid,
directly connected to PC LPT port.

the software measured the refresh of each cell, what is proportional
to the light
the author of that software had some "photos" taken with DRAM online
too.
and eh it wasnt me. but one of my first self-made homecomputers used
the same military gold-ceramic packaged DRAMs

Antti

 

[msg]

Antti wrote:

eh, there used to be a special software that allows a DRAM to work as
black white camera

it was using 64K x 1 military DRAMs, with REMOVED top lid,
directly connected to PC LPT port.

the software measured the refresh of each cell, what is proportional
to the light
the author of that software had some "photos" taken with DRAM online
too.
and eh it wasnt me. but one of my first self-made homecomputers used
the same military gold-ceramic packaged DRAMs

Please see the thread entitled: "Poor Man's image sensor; Was: Re: Simple
Still Camera components?" from June of this year (in comp.arch.embedded).

If anyone wants it and cannot find it, I have the archive of this
project's code and documentation (in German, filename: kuckuck.zip).

Regards,

Michael

 

[sendthis]

Here is a free suggestion (the price is right):
I would write a specific word-pattern with an even mix of 1 and 0 into
every location in the whole DRAM.
Then read back sequentially at a slow pace through all addresses,
always checking the readback.
Sooner or later, you will pick up an error, becaue you exceeded the
refresh delay.
You may want to repeat this with different starting addresses and with
different word patterns.

The problem is during the read (I'm assuming you mean by disabling the
refresh altogether and relying solely on the refresh after read) is
that it takes several seconds to read from the DRAM. This will always
exceed the refresh time right? From the start_address to end_address
it takes quite a while for a 64Mbit DRAM. The spec calls for a 64ms
refresh.