A study from Arthur D Little almost went unnoticed in our holiday inbox. On first read, it is so full of dense mathematical equations that our eyes bled…..
But on persevering, one fascinating factoid can be winkled out. Whilst showing that Ku band has no meaningful advantages in quality any more with Ka band technology marching on, it’s the cost differences which amaze.
Satellites cost about the same per ton to build and launch. ADL is saying that Ka band gets nearly four times as many Megabytes per carrier as Ku band. So a customer wanting an IP service in $ per Mb per month will find that a Ku band service priced like for like is 400% more expensive than Ka band. SatOps rather unhelpfully obfuscate all this by talking about the prices their capacities command in analog – i.e. MHz per month but it seems that- in the new IP digital world, the customer SEES a Mb per month price and this is where the Ka HTS boys like Viasat will blow Ku HTS boys like Intelsat out of the water.
Secondly the Ku HTS services will have to use bigger dishes which are twice the cost and obviously more cumbersome and unsightly. Interestingly one 3G techie told us last year that most mobile masts will not accommodate a dish of more than 1 metre for wind shear and mass reasons – So the Ku HTS boys are OUT of the cellular backhaul market on this basis.
Finally a Ku HTS system has a tiny amount of capacity per country compared to Ka HTS. So If you were a South American country wanting a national broadband deployment, you would very quickly run out of capacity if you chose Ku HTS rather than the Ka HTS services that are on the way on the continent.
An Intelsat EPIC has 10 GHz on board and costs $420m. A mid-range Ka sat has 25 GHz on board and costs $280m. SO, EPIC costs $233 per MHz per month and Ka costs $62 per MHz per month. If now we assume some margin for error in those bits per hertz numbers (we are not sure we are sharp enough to understand the maths there so best to err on the safe side) and assume that Ka gets just double the Mb per MHz that Ku HTS does, then a Ku Mb is SEVEN times more expensive than a Ka band Mb. Plus the Ku dishes are double the cost of Ka dishes.
So why does Ku HTS exist? It is clearly not competitive, if you wanted HTS you would not start there because Ku really is HTS-Lite. Well, we don’t all start with a clean sheet of paper. Intelsat have hugely cash generative long term Ku band contracts with TV companies and others, and they simply cannot afford to invest in new technologies at scale when they also MUST provide re-investment for continuity of service. So whilst they are adding replacement Ku band and trying to groom traffic from several smaller satellites onto one larger satellite in an attempt to save money to stem the interest outflow, they have put some slightly-higher capacity Ku spots on board to try to give something slightly different to their existing customers.
They won’t be winning new business against Ka with this on price or capability, but they might defend existing business better.