Fast asynchronous SRAM output HIGH voltage problem

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
lock attach
Attachments are accessible only for community members.
Anonymous
Not applicable

Hi,

For my application I need a fast asynchronous single port SRAM(organisation: 8bit x32k to 512k x8bit; speed: time<80ns) , which I would  supply with 5V. The min. Output HIGH voltage of the SRAM must be higher than 3,25V.

In this application note (http://www.cypress.com/file/38176/download ) I have found two different RAMs(CY7C199CN, CY7C1049BN), that would  theoretically  fit for my case, because these are devices from the older generation( 250nm & 350nm technologie). According to this application note, these SRAMs should have an Output HIGH Voltage VOH, which is  5V TTL compatible.

But after I have read the ratings from the datasheet, I was a bit confused because there is written a min. Output HIGH Voltage of  VOH=2,4V.

Is the minimum output HIGH voltage VOH>3,25V, if the chip(CY7C199CN or CY7C1049BN or CY7C1049GN) is supplied with Vcc=5V and the output HIGH current (IOH)is  very low (µA)?

It’s true that, according to the ratings in the CY7C1049GN datasheet Vcc=min=4,5V, and the minimal Output HIGH Voltage VOH is more than Vcc-0,5V=4,5V-0,5V=4V?

Or do there exist other SRAM that would be suitable for my application?

The datasheets and the application note you can find in the attachment!

Sorry for my English…

Thanks!

Best regards

0 Likes
1 Solution
nileshbadodekar
Employee
Employee
5 sign-ins 10 replies posted 5 replies posted

Andreas,

Go ahead and use CY7C1049GN devices for 5V legacy systems.

As Noriaki mentioned, these devices can support VOH of upto Vcc-0.5V and thats something we have tested during characterization.

One correction about rise time, this device is a 10ns device, so its very much capable of performing 10-14 ns cycles (as your board load conditions might vary, you have to have some margin)

But I see that the cycle times you are targeting is 80ns, so this should not be a problem.

Have you considered using Micropower SRAMs? The cycle time requirement is 45-50ns and they have very low power consumption

(CY62147G is an example device)

Regards,

Nilesh

View solution in original post

3 Replies
NoriTan
Employee
Employee
25 sign-ins 5 questions asked 10 sign-ins

It is true as the CY7C1049GN datasheet said that VOH(min)=4.0V at IOH=-0.1mA

Please note that this is guaranteed by design and not tested.

By the way, IOH=-0.1mA is enough to drive the data bus?

The minimum capacitance of an output pin is COUT=10pF and the bus will be driven from 0V to 4V.  The rising time is calculated by

tR=COUT*VOH(min)/IOH=400ns

It is too big comparing to the required access speed 80ns.

Does this condition satisfy your requirement?

Regards,

Noriaki

0 Likes
Anonymous
Not applicable

Yes, IOH=-0.1mA is enough to drive the data bus!

For me, the read cycle time is not important.

I forgot to write, that one write cycle should not last more than 80ns.

I think with the CY7C1049GN its not a problem, because the write cycle time in the datasheet is indicated with twc=10ns.

Best regards

0 Likes
nileshbadodekar
Employee
Employee
5 sign-ins 10 replies posted 5 replies posted

Andreas,

Go ahead and use CY7C1049GN devices for 5V legacy systems.

As Noriaki mentioned, these devices can support VOH of upto Vcc-0.5V and thats something we have tested during characterization.

One correction about rise time, this device is a 10ns device, so its very much capable of performing 10-14 ns cycles (as your board load conditions might vary, you have to have some margin)

But I see that the cycle times you are targeting is 80ns, so this should not be a problem.

Have you considered using Micropower SRAMs? The cycle time requirement is 45-50ns and they have very low power consumption

(CY62147G is an example device)

Regards,

Nilesh