Discussion:
[Hackrf-dev] RF path: replacing MAX5864 with MAX5865
Frank Liu
2017-09-19 01:58:19 UTC
Permalink
Hey folks,

We're hoping to extend the existing HackRF design to support a wider
bandwidth. Looking at datasheets, it seems that we can simply replace the
MAX5864 (22Msps) with a MAX5865 (40Msps) with pretty much no change in
hardware design.

Was wondering if anybody had tried this already. Our HackRF will be
"standalone", meaning that we won't be constrained by USB2.0's I/O speed.

Thanks!
Michael Ossmann
2017-09-19 17:39:08 UTC
Permalink
Frank,

Yes, you can replace MAX5864 with MAX5865. However, you may still run into
bandwidth limitations from the MAX2837 (31 MHz maximum baseband filter
bandwidth) and SGPIO (our implementation tops out at about 25 Msps).

Michael
Post by Frank Liu
Hey folks,
We're hoping to extend the existing HackRF design to support a wider
bandwidth. Looking at datasheets, it seems that we can simply replace the
MAX5864 (22Msps) with a MAX5865 (40Msps) with pretty much no change in
hardware design.
Was wondering if anybody had tried this already. Our HackRF will be
"standalone", meaning that we won't be constrained by USB2.0's I/O speed.
Thanks!
_______________________________________________
HackRF-dev mailing list
https://pairlist9.pair.net/mailman/listinfo/hackrf-dev
Frank Liu
2017-09-20 00:53:56 UTC
Permalink
Thanks for the quick reply! For the MAX2837, we were actually thinking of
just disabling the baseband filter via the LPF register. We'll be operating
in transmit mode for most of the time anyway, and we can apply digital
filters to any received signals we need to clean up.

--Frank
Post by Michael Ossmann
Frank,
Yes, you can replace MAX5864 with MAX5865. However, you may still run into
bandwidth limitations from the MAX2837 (31 MHz maximum baseband filter
bandwidth) and SGPIO (our implementation tops out at about 25 Msps).
Michael
Post by Frank Liu
Hey folks,
We're hoping to extend the existing HackRF design to support a wider
bandwidth. Looking at datasheets, it seems that we can simply replace the
MAX5864 (22Msps) with a MAX5865 (40Msps) with pretty much no change in
hardware design.
Was wondering if anybody had tried this already. Our HackRF will be
"standalone", meaning that we won't be constrained by USB2.0's I/O speed.
Thanks!
_______________________________________________
HackRF-dev mailing list
https://pairlist9.pair.net/mailman/listinfo/hackrf-dev
Michael Ossmann
2017-09-20 00:56:33 UTC
Permalink
Frank,

I've tried without success to disable the baseband filter. If it works for
you, please let me know how you did it.

Thanks,

Michael
Post by Frank Liu
Thanks for the quick reply! For the MAX2837, we were actually thinking of
just disabling the baseband filter via the LPF register. We'll be operating
in transmit mode for most of the time anyway, and we can apply digital
filters to any received signals we need to clean up.
--Frank
Post by Michael Ossmann
Frank,
Yes, you can replace MAX5864 with MAX5865. However, you may still run into
bandwidth limitations from the MAX2837 (31 MHz maximum baseband filter
bandwidth) and SGPIO (our implementation tops out at about 25 Msps).
Michael
Post by Frank Liu
Hey folks,
We're hoping to extend the existing HackRF design to support a wider
bandwidth. Looking at datasheets, it seems that we can simply replace the
MAX5864 (22Msps) with a MAX5865 (40Msps) with pretty much no change in
hardware design.
Was wondering if anybody had tried this already. Our HackRF will be
"standalone", meaning that we won't be constrained by USB2.0's I/O speed.
Thanks!
_______________________________________________
HackRF-dev mailing list
https://pairlist9.pair.net/mailman/listinfo/hackrf-dev
Frank Liu
2017-09-20 01:18:27 UTC
Permalink
Wow, that is a bit of a bummer. Just to make sure we're on the same page,
I'm referring to page 20 of the MAX2837 datasheet:
https://datasheets.maximintegrated.com/en/ds/MAX2837.pdf. Is it just
straight up wrong?
Post by Michael Ossmann
Frank,
I've tried without success to disable the baseband filter. If it works for
you, please let me know how you did it.
Thanks,
Michael
Post by Frank Liu
Thanks for the quick reply! For the MAX2837, we were actually thinking of
just disabling the baseband filter via the LPF register. We'll be
operating
Post by Frank Liu
in transmit mode for most of the time anyway, and we can apply digital
filters to any received signals we need to clean up.
--Frank
Post by Michael Ossmann
Frank,
Yes, you can replace MAX5864 with MAX5865. However, you may still run
into
Post by Frank Liu
Post by Michael Ossmann
bandwidth limitations from the MAX2837 (31 MHz maximum baseband filter
bandwidth) and SGPIO (our implementation tops out at about 25 Msps).
Michael
Post by Frank Liu
Hey folks,
We're hoping to extend the existing HackRF design to support a wider
bandwidth. Looking at datasheets, it seems that we can simply
replace the
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
MAX5864 (22Msps) with a MAX5865 (40Msps) with pretty much no change
in
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
hardware design.
Was wondering if anybody had tried this already. Our HackRF will be
"standalone", meaning that we won't be constrained by USB2.0's I/O
speed.
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
Thanks!
_______________________________________________
HackRF-dev mailing list
https://pairlist9.pair.net/mailman/listinfo/hackrf-dev
Michael Ossmann
2017-09-20 14:21:39 UTC
Permalink
I've observed no change of behavior when setting LPF disabled in register 2.
Specifically I have looked for aliasing.
Post by Frank Liu
Wow, that is a bit of a bummer. Just to make sure we're on the same page,
https://datasheets.maximintegrated.com/en/ds/MAX2837.pdf. Is it just
straight up wrong?
Post by Michael Ossmann
Frank,
I've tried without success to disable the baseband filter. If it works for
you, please let me know how you did it.
Thanks,
Michael
Post by Frank Liu
Thanks for the quick reply! For the MAX2837, we were actually thinking of
just disabling the baseband filter via the LPF register. We'll be
operating
Post by Frank Liu
in transmit mode for most of the time anyway, and we can apply digital
filters to any received signals we need to clean up.
--Frank
Post by Michael Ossmann
Frank,
Yes, you can replace MAX5864 with MAX5865. However, you may still run
into
Post by Frank Liu
Post by Michael Ossmann
bandwidth limitations from the MAX2837 (31 MHz maximum baseband filter
bandwidth) and SGPIO (our implementation tops out at about 25 Msps).
Michael
Post by Frank Liu
Hey folks,
We're hoping to extend the existing HackRF design to support a wider
bandwidth. Looking at datasheets, it seems that we can simply
replace the
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
MAX5864 (22Msps) with a MAX5865 (40Msps) with pretty much no change
in
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
hardware design.
Was wondering if anybody had tried this already. Our HackRF will be
"standalone", meaning that we won't be constrained by USB2.0's I/O
speed.
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
Thanks!
_______________________________________________
HackRF-dev mailing list
https://pairlist9.pair.net/mailman/listinfo/hackrf-dev
Frank Liu
2017-09-20 15:17:24 UTC
Permalink
I see. Many thanks for all your help.

Hopefully my last question - were you able to fiddle with the filter mode
(bits D3:D2, LPF register 1) i.e TX LPF vs RX LPF? My thinking is that if
we configure the filter to operate for RX only we may still be able to at
least transmit a higher BW signal without requiring a different chip.
Post by Michael Ossmann
I've observed no change of behavior when setting LPF disabled in register 2.
Specifically I have looked for aliasing.
Post by Frank Liu
Wow, that is a bit of a bummer. Just to make sure we're on the same page,
https://datasheets.maximintegrated.com/en/ds/MAX2837.pdf. Is it just
straight up wrong?
Post by Michael Ossmann
Frank,
I've tried without success to disable the baseband filter. If it
works for
Post by Frank Liu
Post by Michael Ossmann
you, please let me know how you did it.
Thanks,
Michael
Post by Frank Liu
Thanks for the quick reply! For the MAX2837, we were actually
thinking of
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
just disabling the baseband filter via the LPF register. We'll be
operating
Post by Frank Liu
in transmit mode for most of the time anyway, and we can apply
digital
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
filters to any received signals we need to clean up.
--Frank
Post by Michael Ossmann
Frank,
Yes, you can replace MAX5864 with MAX5865. However, you may still
run
Post by Frank Liu
Post by Michael Ossmann
into
Post by Frank Liu
Post by Michael Ossmann
bandwidth limitations from the MAX2837 (31 MHz maximum baseband
filter
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
Post by Michael Ossmann
bandwidth) and SGPIO (our implementation tops out at about 25
Msps).
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
Post by Michael Ossmann
Michael
Post by Frank Liu
Hey folks,
We're hoping to extend the existing HackRF design to support a
wider
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
bandwidth. Looking at datasheets, it seems that we can simply
replace the
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
MAX5864 (22Msps) with a MAX5865 (40Msps) with pretty much no
change
Post by Frank Liu
Post by Michael Ossmann
in
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
hardware design.
Was wondering if anybody had tried this already. Our HackRF will
be
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
"standalone", meaning that we won't be constrained by USB2.0's
I/O
Post by Frank Liu
Post by Michael Ossmann
speed.
Post by Frank Liu
Post by Michael Ossmann
Post by Frank Liu
Thanks!
_______________________________________________
HackRF-dev mailing list
https://pairlist9.pair.net/mailman/listinfo/hackrf-dev
Loading...