[Date Prev][Date Next][Date Index]
V/F converter
(I sent this message last week, but our mailer deamon told me something
went horribly wrong in the transmission. Sorry if this is a repeat
for you.)
Several weeks ago, I described briefly, and asked for comments on, a
voltage/frequency converter quote I received from Nova R&D Inc. I got
good ideas from several sources, notably Mark Rivers mention of Peter
Siddons' ideas implemented by P. J. Pietraski (NIM A319 (1992)414).
(Thanks all.) Nova has quoted a cost-reduced, independent
implementation of those ideas. Excerpts:
NIM four-channel V/F converter
------------------------------
converter: AD652A (-B, -J, -K, PLCC version optional)
input amp: AD624x (optional AD625x)
gain: x1, x100, x200, x500
(x10, x1000 optional, requires AD625x $20/module)
input: 0 to +/- 1V, 0 to +/- 10V
(0 to +/- 5V optional, requires PLCC version of AD652 $24/module)
(impedance unspecified, BNC or LEMO)
output: 0 to 1MHz.
open-collector TTL to BNC connector
(options: CMOS $?, 50-ohm $20, LEMO $36)
nonlinearity: AD652A: 0.002% typical, 0.02% max. (.005% max for -B)
AD624A: 0.005% max (lower for -B and -C)
response time: one period of new output frequency plus one clock period
(I'm pretty sure the clock is 2MHz)
delivery: prototype: 90 days
production: 60 days after prototype is accepted
nonrecoverable engineering charge: $6.8k including first prototype.
quantity price
---------------------
1 1800
6 1200
12 1000
25 800
My understanding is that quantity pricing is cumulative, so that if
somebody buys 12 modules, everybody afterward gets them at $1000 ea.
SRI-CAT will probably get 12, so if other CATs can come up with a total
of 15, we can just about bury the NRE charge.
Let me know if you want a copy of the quote. If you want to talk to
Nova directly, speak with Tumay Tumer at (909)781-7332.
I don't know the impact of different option packages on quantity
pricing, but I can't imagine it's good. I recommend we agree on the
PLCC option ($24/module) that gives an additional 0 to +/- 5V range; I
can live with BNC connectors (LEMOs would cost $36/module); I want the
x10 gain option ($20/module). I'm guessing the best-nonlinearity
chips will add an additional $100/module.
Preferences?
Tim Mooney