Question : Dual Microsoft DHCP scope configuration.

I've been reviewing configuration advice for dual DHCP configuration.  I see the typical 80/20 solution, but I'm wondering if someone has real world experience with that configuration.  I have a large 24x7 production facility that I cannot afford to make mistakes on.  I currently backup the DHCP configuration to another AD controller, but that requires manual intervention to enable that scope if the primary DHCP server is offline.

The typical instructions for the 80/20 DHCP config, say to split the scope with no overlap, but to me that seems wrong.  It seems like they should have the same scope defined, but manipulate the split by using excluded ranges.  
Example -  DHCP 1   scope 10.0.0.100 - 10.0.0.200/24   Exclusion of 10.0.0.181-10.0.0.200
                 DHCP 2   scope 10.0.0.100 - 10.0.0.200/24   Exclusion of 10.0.0.100 - 10.0.0.180

In the event of an extended outage for the primary DHCP scope (such as hardware failure),
you could simply remove the exclusion range and the DHCP 2 server would run the full scope.
When the primary DHCP returns online, simply reinstate the original exclusion range on DHCP 2.  
Of course both DHCP servers would be configured with conflict detection.  My L3 switches have ip helper statements for both servers. Primary is listed first.
 
To me it seems like a better solution. Am I leaving out something that would not allow this to work?

Answer : Dual Microsoft DHCP scope configuration.

For me it depends if you will get the faulty hardware up before the leases all drop off (the scope of that server)

As a client will hold the DHCP ip for the lease time (checking twice unicast to the assigning DHCP server, assuming no reply it will hold off till the final 100% completion of the lease. At this point it would request a lease broadcast and get the 20% DHCP server scope which should be fine,

I can see your point but if your 24/7 you would hopefully be looking at 4 hour min response on server failure so you should be fine in an 80/20 rule, Your exclusions being switched on and off in event of failure will allow you more flexibility in recovery time but I feel it will cause you headaches when you switch back to normal 80/20 service even with collision detection.