Designer’s Guide Community

Pages: 1

High speed, high mirror ratio current mirror (Read 3693 times)

AnalogDE Senior Member Offline Posts: 137	High speed, high mirror ratio current mirror May 27^th, 2008, 2:24pm I am trying to design a current mirror with a high mirroring ratio (1:12, input current=40uA, output=480uA). �Do you think the topology attached could work? �How should I size the source follower and current source? �I need very fast settling time, ~10ns. �
Back to top	CBUF.GIF IP Logged

weber8722 Community Member Offline Posts: 95	Re: High speed, high mirror ratio current mirror Reply #1 - Mar 17^th, 2009, 7:27am Hi, for good settling you need a frequency compensation. Use a series RC from input to Vdd, like 500fF + 10k. Otherwise you get a large overshoot and big noise peaking too. Bye Stephan
Back to top	IP Logged

rf-design Senior Member Offline Reiner Franke Posts: 165 Germany	Re: High speed, high mirror ratio current mirror Reply #2 - Mar 24^th, 2009, 10:39am Very fast depending on your tech, Vdsat and Length. A good guess is to have a sqrt(12) * input current on the mirror buffer.
Back to top	IP Logged

vivkr

Community Fellow

Offline

Posts: 780

Re: High speed, high mirror ratio current mirror
Reply #3 - Mar 25^th, 2009, 12:04am

AnalogDE wrote on May 27^th, 2008, 2:24pm:

I am trying to design a current mirror with a high mirroring ratio (1:12, input current=40uA, output=480uA). Do you think the topology attached could work? How should I size the source follower and current source? I need very fast settling time, ~10ns.

I would also try to see if a Wilson current mirror could be used (headroom permitting). This is generally quite fast for high mirror ratios.

Regards,

Vivek

IP Logged

Pages: 1

‹ Previous topic | Next topic ›