Here are some 2510 notes and ideas sent to me by Donald Gruneisen KM6ZQ (edited for clarity and readability).  Our exchange ended up being too lengthly to put on the main HR2510 page, so I decided to create a new page.  Don works for Motorola, and as we were corresponding, he was working on the Iridium de-orbit plan!

This under construction, so it's still pretty rough.

From Don:

i just came across your 2510 info page...nice job! i do a search for
2510/2600 every now and then, just to see what kind of new garbage mods are
showing up, and was surprised, delighted even, to see some decent
information for a change.

i have worked through much of what you presented myself, comparing and
swapping between 2510s and HTX100s. done the crystal filter swap (with
carrier osc shift), added LSB to an HTX, RIT range reduction, mixer balance
pots, etc. glad to see someone took the time to write this stuff up!

i'm also a fan of the chipswitch. roger, dan, and darryl are a great bunch
of guys. bit buggy, though, huh? ever lose power during TX? you get to
reprogram everything!  i first noticed it when i blew a fuse, while
transmitting of course, and had the radio reset upon replacing it. later i
tried it on purpose to try to isolate and correct it. never had any luck,
even after questioning chipswitch on several occasions. now i find its an
easy way to clear the memories...just push the TX button in and flip the
power switch off. once or twice should be enough to get it! easier than the
master reset.

i have a few additions for you to think about...maybe you already have. ever
unlock the RIT but find that the carrier osc shifts ~100 Hz between TX and
RX, but only on LSB? this problem is worst on a 2510 as i think the 2600 and
the HTX had some minor revisions to the osc circuit which helped. you can
compare schematics and make all the changes with some improvement, or i've
seen people de-Q the LSB tank by paralleling a 1k resistor across the
adjustment coil or the choke that feeds the diode to select that coil. i
looked a little closer and found that the Vcc to the osc transistor changes
between TX and RX, because of going through a different number of diodes for
each. this causes the bigest problem on LSB because the XTAL is being pulled
farther. ended up clamping the supply to 5.1 volts with a zener at the
collector which fixed the problem and improved overall stability.

you may also know that the firmware in the UC1201 (2510) is the same as the
UC1208 (HTX). one pin (i'll have to look it up) tells the chip which radio
it is in. funny things happen when you change the logic level of that pin!
like the HTX shifts the PLL 2.5 kHz instead of 1.5 kHz (i finally found
someone who will understand this!) and the 2510 displays "u" or "c" in the
band display, the only modes on an HTX, right? well, how about using the HTX
memory functions on a 2510...can it be done? YES, but it is not for the
typical "use vice grips to remove the ALC transistor" types. find where the
memory key would go on a 2510 key matrix (with the help of the HTX
schematic) and add a switch. now you have to devise a way to switch to HTX
mode, then use the memory switch to find (or set) the desired channel, and
switch back to 2510 mode. problem is, any time you switch modes you need to
reinitialize the CPU in order for it to take properly. no problem when going
to HTX mode, just reset it. going back to 2510 though, resetting loses the
channel you just pulled out of memory! the trick: carefully control the stop
line to the processor. sounds like a job for a PIC, but i never got that
far.

how about 6 meters? the frequency plan is perfect for it, just switch to low
side LO instead of high side and you are close. build new RX and TX filters
and modify the front end and PA and you're set. i use the MRF497, but the
477 will still work at 50 MHz. did you know hat some of the chips have 6
meters in them? the display and N counters are all set for 50-54. one of my
radios is essentially a kenwood TS-60 for a fraction of the price! i wish i had
a solid formula. you need to start with a UC1251 from a
2600. as far as i know, it is the only one with this function. the PA is
pretty easy to get going, and the receiver front end will sort of work as
is. the hard part, for me anyway, is the TX and RX filters. you know, the
ones that say "do not adjust." i tried rewinding them, but finally replaced
them completely with lightly coupled resonant tanks. they won't cover much
more than 1-2 MHz though, and my implementation is pretty ugly. this part
needs some more attention. maybe i can get it back on the bench soon.

ever see any difference in operation based on 2510, 2600, or lincoln
settings? i thought the N-codes might be different, since the 2510 service
manual shows strange frequencies on its block diagram. i guessed maybe
earlier versions (lincoln mode?) used a different frequency scheme but they
don't change between modes. the N-count should change between TX and RX in
CW on a 2600, since i think the offset is different on that model, but i
never verified a change with the chipswitch in different modes. i have a few
(different versions of the) chipswitch, including the old password 15
meter ones. ever try to get one to go that low? not worth it, IMHO. but
different N-code approaches may be better for this than others. how
successful have you been at getting to 24.8? here's a tip: replace the 7808
on the VCO supply with an '09 or '10 for a bit more tuning range.
the early versions only went to 25.5 unless you put in a unique password to
unlock the lower frequencies. then they would dial all the way to 20.5, but
good luck getting them to work there! check
http://www.chipswitch.com/Password.htm if you haven't seen it before. there
is also a way to get dan v's name to scroll across the display but i'll have
to look that up, too. i have one of the early chips left to play around with
(or copy) someday. this is what sparked my curiosity about the N-codes in
the different uniden chips. on the old chipswitches, below about 25.5 MHz
they start decrementing the 110 MHz loop instead of the main loop. i don't
know why, it may be a limitation of the PLL0305 or something, but basically
it means you need to broaden both VCOs instead of just one. the whole deal
never had much promise anyway, so they just fixed them all at 24.8.

how successful have you been at getting to 24.8? here's a tip: replace the
7808 on the VCO supply with an '09 or '10 for a bit more tuning range.
i've seen an older lincoln that seemed to
have a less sensitive VCO that changing to a 9V regulator (and tuning closer
to 8V at 29.9) fixed. i also tried going to 10V when i was working on the 15
meter expansion.

(Re: Low RF sensitivity on 12 Meters)
the RX sensitivity problem could be from your PLL board. check the output
filter and make sure it's not dropping the level too much down there. i seem
to recall having to increase the coupling between the resonators or
something? maybe i'm dreaming. the RX mixer, being discrete, is probably
more sensitive to low LO drive than the IC TX mixer.
 

(Re: FM RX performance)
 i don't recall FM alignment instructions in the service manual, but
there is one adjustment shown at
http://members.tripod.com/~Malzev/cb-radio/hr2600.htm and a lot of other
good info, too. i thought the only adjustable coil on the FM board was the
quadrature coil, but maybe an adjustment would help. the demod chip is (a
clone of) our MC3357 which is a pretty decent part, plus you get double
conversion, which should help. there are a few different FM boards, some
with a buffer amp and some without. i don't know which is better, but if i
had to guess... maybe the 1st filter (xtal) is a little bit narrow?