Maxtrac 800 to 902 conversion advice needed

ka4ymy · Post by **ka4ymy** » Thu Jan 04, 2007 4:57 pm

Hi guys..

I'm working on replacing the repeater receiver on my 902 ham system, and like a lot of you I'm looking to use a converted Maxtrac.

I have read from a lot of you that you're achieving "excellent" receive sensitivity, however, I haven't seen actual numbers. The one I have in front of me now has been modified for the VCO switching, 915 filters installed, and programmed up for my repeater input of 902.7125.. The best I'm seeing here is around .5uv for 12db sinad.. I consider that far from excellent. My present repeater receiver is a Hamtronics crystal strip, and I'm nabbing well over .2uv on it.. The Hamtronics however, has a rather poor front end and my site is a noisy site. The Maxtrac would no doubt be more selective and cleaner, but I would like to get a bit more out of this than .5uv if you know what I mean!

So what I'd like to know is, what sort of sensitivity are you guys getting on your Maxtracs?? Is mine about in line with everyone else's, or is mine a bit deaf?

Thanks guys for your input,
Steve

kcbooboo · Post by **kcbooboo** » Thu Jan 04, 2007 5:14 pm

It looks like you converted an 800 MHz MaxTrac to receive 902. Did you test the radio's sensitivity in its original configuration first? If it was poor before you converted it, then you have to deal with that first. What brand of 915 MHz filters are you using?

I've seen several different methods of replacing the front end filters, so perhaps your technique isn't optimal. I completely unsolder the shield from under the first filter, then completely unsolder both filters and remove them from the RF board. I clean the board with alcohol and inspect for any damage. Then I install new TOKO 6DFB-915E-10 filters, short out Q205, replace the front end shield, and go to work on the logic board (firmware, blank, initialize). I've always gotten excellent results, always under 0.4uV for 20dB quieting (no crackles in the signal), typically even better than that.

I do know that some people remove the filter body from the mounting clip and leave the clip on the RF board. Then they pull the new clip off the new filter and install the new filter into the old clip on the board. I had several that were done that way and every one measured over 1uV for 20dB quieting at 902 MHz. I ended up pulling all of the filters out completely and putting in brand new TOKO filters my way; that fixed every board.

I don't know what the bandwidth is on the MuRata 915 MHz filters. They may not be as wide as the TOKO filters. I do know that the stock 938 MHz MuRata filters were only about +/- 8 MHz wide; sensitivity at 927 MHz on a 900 MHz MaxTrac is poor, a lot worse than at 935 MHz. If the filters you're using aren't wide enough, then that would give you reduced sensitivity at 902.

You can try shorting out the filters to see if you get any improvement. The receiver may not be as selective but it shouldn't be a problem if you run DPL on your repeater input. You should see excellent sensitivity even with the original 860 MHz filters the radio came with, once they're shorted out.

Bob M.

ka4ymy · Post by **ka4ymy** » Thu Jan 04, 2007 7:49 pm

Hi Bob--

I've seen quite a number of your posts both here as well as AR902.. Thank you for offering your knowledge to so many of us out here across the fruited plain

I did test the original sensitivity, it was around .35uv typical (no crackles in the signal). I initially tried placing jumpers across the filters after I removed them from the board, just to see what I could get out of it as it was. I had the .5uv at 902.7125.. I proceeded to install a set of Murata 915 filters.. Same level of sensitivity. So I'm not sure if what I'm experiencing is a set of lossy filters. It seems more like the board is losing sensitivity towards the high end like this. Something else I have not done is change the firmware. This is a Maxtrac 840. I simply programmed a single conventional channel into the radio. Could the firmware have anything to do with poor sensitivity?

Steve

W4WTF · Post by **W4WTF** » Thu Jan 04, 2007 8:16 pm

ka4ymy wrote:Hi Bob--

I've seen quite a number of your posts both here as well as AR902.. Thank you for offering your knowledge to so many of us out here across the fruited plain

I did test the original sensitivity, it was around .35uv typical (no crackles in the signal). I initially tried placing jumpers across the filters after I removed them from the board, just to see what I could get out of it as it was. I had the .5uv at 902.7125.. I proceeded to install a set of Murata 915 filters.. Same level of sensitivity. So I'm not sure if what I'm experiencing is a set of lossy filters. It seems more like the board is losing sensitivity towards the high end like this. Something else I have not done is change the firmware. This is a Maxtrac 840. I simply programmed a single conventional channel into the radio. Could the firmware have anything to do with poor sensitivity?

Steve

I didn't know you could do this without converting them to conventional and such... so you can swap filters and program a conventional channel and it works (or works as well as what you have so far)?

That will save mea little time and money if so!

ka4ymy · Post by **ka4ymy** » Thu Jan 04, 2007 8:30 pm

KF4PEP--

Well I guess my take on it is why wouldn't it work?? Perhaps I am totally missing the entire point of changing out the firmware in this radio, but if the radio is already capable of programming conventional channels into it, then why change the firmware?? OK, so it will do trunking as well as conventional.. I simply ignored the trunking part and programmed 902.7125 receive, and entered the word "blank" into the transmit entry, and voila! I have a somewhat deaf 902 single channel radio. Unless there is something related to blanking the board and re-programming the logic in the unit. I suppose there could be some weird tuning data that the radio needs in order to more properly align itself to the 902 range.

No doubt this is an excellent way to obtain a very good repeater receiver, I just wonder is there more to this than meets the eye..?

Steve

W4WTF · Post by **W4WTF** » Thu Jan 04, 2007 9:29 pm

ka4ymy wrote:KF4PEP--

Well I guess my take on it is why wouldn't it work?? Perhaps I am totally missing the entire point of changing out the firmware in this radio, but if the radio is already capable of programming conventional channels into it, then why change the firmware?? OK, so it will do trunking as well as conventional.. I simply ignored the trunking part and programmed 902.7125 receive, and entered the word "blank" into the transmit entry, and voila! I have a somewhat deaf 902 single channel radio. Unless there is something related to blanking the board and re-programming the logic in the unit. I suppose there could be some weird tuning data that the radio needs in order to more properly align itself to the 902 range.

No doubt this is an excellent way to obtain a very good repeater receiver, I just wonder is there more to this than meets the eye..?

Steve

I was thinking along those same lines, and wondering why all the extra was needed.....

Did you enter the freq direct, or put in 812.5125?

Hmmm time to experiment.

kcbooboo · Post by **kcbooboo** » Fri Jan 05, 2007 3:49 am

If the radio will already accept conventional modes, there's no need to replace the firmware, blank, and initialize the radio to be conventional only, especially if it's only going to be used to receive one frequency. As long as the radio can't power up on some other mode, you're good to go.

There was someone else on here who had some specs on MuRata filters. As long as they're centered at 915 MHz and have a bandwidth of at least +/- 13 MHz, they should be as good as the TOKO filters, and you should see good sensitivity.

I'm surprised that you lost so much merely by shorting out the filters. That was not my experience. The filters themselves do have a bit of loss, and I found that I actually had better sensitivity with them shorted out. They were still in place, however, so maybe by removing them and shorting them out you end up losing something.

I don't think I mentioned it earlier, but I also clean the back of the RF board after soldering in the new filters. Sometimes the flux is more conductive than I care to have it; besides it looks better afterwards. Everything has to be soldered well. The ground tabs on the filters need to get hot enough so the solder flows around them on both the top and bottom of the board. They should look like the originals when the job is done.

I hate to say the MuRata filters have different specs because I have never seen them, but it's certainly possible. Jeff W6JK and I have sold hundreds of the TOKO filters and used dozens ourselves and have had no problems at all, on both 800 and 900 MHz MaxTracs; several other people have also done that much business and if people were getting such lousy sensitivity the word would have gotten around pretty quickly.

Bob M.

Max-trac · Post by **Max-trac** » Fri Jan 05, 2007 7:23 am

I think most poeple "blank and make conventional" because you might want a 32ch radio, or you have a 2ch that is trunking only or you are swapping logic boards etc. And once you are set up and have done it a couple of times, it is easy.

I would at least make sure it is on freq.
And then maybe do a complete alignment, but if it was good on 800 before, and it is at least working on 902, then I suspect it is just too far out of its design spec and each radio is a little different in the way they behave.

ka4ymy · Post by **ka4ymy** » Fri Jan 05, 2007 9:09 am

OK guys--

Thank you again for your advice and ideas.. This is what makes projects like this interesting; folks that are willing to lend a hand because they've been there before. It makes a difference!

Alright, I sat down with this Maxtrac and decided to take Bob's advice to heart; re-visit the method by which I put the filters in ,VCO mod etc.. First off, I initially shorted out the base to emitter junctions on the VCO bandswitch transistor. The transistor has now been completely removed from the pc board. Secondly, I un-installed the Murata filters.. cleaned the board and checked everything under a magnifying glass. I then proceeded to lay silver wire in place of the filters, also using silver solder so as to provide the lowest loss possible. I also cleaned and re-soldered the bottom shield back on to the board using silver solder as well. I put it all back together, and my result appears to be better. With this configuration, I am now making .4uv with no crackles on receive audio, 1kc test tone at 3kc deviation. At .35uv, I begin to have crackles, and at .25uv, the squelch closes up, but with quite a noisy signal. The quality of the signal drops off fast! Now, to deal with those removed pc-board Murata filters.. I have available a Hyperlink brand 4-pole bandpass filter that I can connect to the radio's antenna connector. It's a 902-928 bandpass only filter, fits squarely in your hand. These can be found on ebay for around $100 brand new. They are very nice and quite rugged. With this filter, plus the duplexer, I honestly believe that this configuration would work. I am giving up around .10 to .15uv of my receive sensitivity at this point in exchange for a receiver that not only sounds better, but also provides improved selectivity than the Hamtronics strip offers. I would prefer to have the filters back on the pc board, however, it's pretty good as it is, I'm afraid to disturb it!!

Do you guys feel that this is getting more in line to what I should expect from the Maxtrac itself?

Steve

kcbooboo · Post by **kcbooboo** » Fri Jan 05, 2007 9:47 am

I've only measured the sensitivity with the stock filters installed but shorted out. For all I know, they are part of the circuit even when shorted, and might provide a better impedance match.

You can easily reset the squelch pot on the RF board; it's the blue pot with "SQ" next to it. As long as the VCO is locking, press the MON button on the front panel or make sure the MON LED is lit, then turn the pot counter-clockwise to open the squelch, and with no signal applied, turn it clockwise until it just closes. At that threshold, the radio should open below 0.1uV. I think I still have an unmodified MaxTrac upstairs; I'll run some tests on it for comparison, as well as one that has had TOKO filters installed.

Shorting the emitter to base of Q205 completely disables it and lets the VCO operate in the higher of two ranges, which is necessary for it to receive around 902 MHz. It's also easy to reverse if you ever want to put the radio back to stock. Now that you've pulled it, wanna bet how long it is until you can't find it?!?!?!

Silver solder and/or wire is definitely an overkill, and shouldn't be necessary.

When you feed a signal to the receiver, you may have to adjust your sig gen frequency a bit for the best signal. If it makes an improvement, the warp setting in the service menu might be off, so the receiver ends up being off-frequency. Since Q205 is missing or disabled, the VCO can only operate in the 850-870 MHz range, and the test frequencies the software uses will probably want to program it down in the 806-825 MHz range. About all you can do is program a frequency of your own choice, such as 860.00000 MHz, key the transmitter, and measure the output frequency. Then adjust the warp setting to get that exactly on the money. That should get the receiver close, but since the 2nd oscillator is not adjustable, that's all you can do.

Bob M.

kcbooboo · Post by **kcbooboo** » Fri Jan 05, 2007 10:54 am

I tested three radios:

stock 800 MaxTrac on 851 MHz
800 MaxTrac with TOKO 915 MHz filters on 902.4 MHz
stock 900 MaxTrac on 927.5 MHz

Note that the 900 MHz MaxTrac was a lot more sensitive in the 935-941 MHz band than down at 927.5 MHz with its stock filters which are not very wide. At some point this radio will have TOKO 915 MHz filters installed; it should then have sensitivity equal to the 800 MHz radio with the TOKO filters.

Bob M.

W6JK · Post by **W6JK** » Fri Jan 05, 2007 7:19 pm

Nice work, Bob! I'm forced to admit that my sensitivity measurements have consisted of verifying that the UUT hears various local repeaters as well as previous units did. My signal generator tops out at 480 MHz. My station has outrun my test equipment. I'm sure glad I bought that 3GHz plug-in for the frequency counter when I found it a while back!

Jeff