Spectra for repeater receiver

[John G]

I seem to remember some discussion about using a Spectra for a repeater receiver, but I can't find it. There is an article on Repeater builder that says it can't be done. Though I don't dispute anything said in that article, I still want to do it. I plan on bypassing the internal ceramic filter and using an external preselector. So, the only two obstacles remaining are getting the software to accept the lower frequency and getting the VCO to go low enough. Has anyone here actually done this? Can you give me some pointers?
Thanks in advance.

[jhooten]

First, what band?

[John G]

900 of course.

[jhooten]

I'm told it is better to start with an 800 spectra and move it up.
And the 800 Maxtrac is even better.

Join the 900 group on yahoo groups lots of experience there.

[John G]

I am a member of several of the 900 groups over there. I do have a 800 Spectra but didn't want to use anything with 25 KHz wide filters. I have converted a couple of Maxtracs but don't want to use them this time. It seems to me that the Spectra is a superior receiver. I have a couple of Tait receiver strips that don't require any conversion but I just wanted to go with a Spectra this time. I also have a PURC transmitter that I can't find anything on. It isn't a PURC5000. I'll post the model if anyone even thinks they might have some info. I do have a PURC book but the model is enough different that it can't be used.

[jhooten]

I'm getting real close to just ordering the stuff from Hamtronics myself.

[John G]

After pulling my hair out trying to keep Hamtronics stuff working, I try to avoid it if at all possible. We had a UHF autopatch link receiver that would work anywhere except at the repeater site. MASTR 2 cured that problem for good.

[motorola_otaku]

The reason the Spectra has such an excellent receiver is because of its preselector network - it uses aggressive bandpass filtering before feeding a preamp (and they aren't ceramic like the Maxtrac, it uses laser-tuned bandpass cavities.) If you go bypassing stuff on there to try to force it down to 902 you are going to end up with a very deaf receiver with a barn-door front end. And this of course assumes you can get the VCO to lock on RX at 902 and nothing more complex than a hex edit will be needed to make the radio take 902 receive frequencies, which very well may not be the case.

If you can't find a MSF5000 and you can't afford a Quantar, stick with a modified Maxtrac. I don't mean to sound rude, but there's a reason no one has tackled what you're trying to do. Or if you really want to apply your experimental energies, get a MTR2000 and hack it for 902/927. Those have been done successfully but the method of doing so hasn't been publicized.

[John G]

I don't know which Spectra you have been looking at but the ones I have use a ceramic filter. It is bigger than the Maxtrac one and has less loss but it is still just a ceramic filter. I did a Google search for Spectra mods and found an article where some fellow says he uses one for a 900 repeater receiver. He says it works well. About my only real concern is that the local oscillator is at the same spot where the RF comes in. I think I will put a circulator there to help block that L.O. power and terminate it. I plan on using a HEMT preamp after a good 4 can preselector. Should be OK. I have done a couple of Maxtracs and there is nothing wrong with them. I also have a couple of Taits I can use as well as assorted Kenwoods. I even have an EFJ already fully converted. I just wanted to use a Spectra because I have a couple lying around and they seem like pretty good receivers. I still have to re locate my RIB and cable as well as the RSS. It will be a while. I'll post results when it finally happens.

[SlimBob]

IMO, it's easier to convert the 800MHz Maxtrac to use as an RX.

The Spectra is ill-suited for use as an RX for two reasons: 1) getting COS out and it's not designed to receive anywhere near 902MHz.

[John G]

I definitely agree that the Maxtrac is easier. I don't need COS because I am going to feed discriminator audio into a repeater controller and I don't think that will be all that hard to find. Getting the RSS to accept 902 frequencies won't be too difficult, I hope. Getting the VCO low enough shouldn't be either. We'll see. My main objection to the 800 Maxtrac is that it is designed for 25 KHz channel spacing, 5 KHz deviation and 900 uses 12.5 KHz spacing and 2.5 KHz deviation. If I wanted the easiest way, I would just program one of the Tait receivers I already have.

[SlimBob]

Oh.

The issue is the IF filters on the Maxtrac and the channel spacing/programming.

That's not an easy puzzle to solve.

This information may help:

http://www.repeater-builder.com/maxtrac ... lters.html
http://www.repeater-builder.com/maxtrac ... o-ext.html

I would imagine that an older radio, like an 800MHz Mastr II might have helicals that could be stretched to cover the band and provide additional selectivity.

RB has some LBIs of 900MHz Mastr IIs. Likely, anyone who had a Mastr II on 900MHz SMR started in the 1980s.

[John G]

The IF filters aren't the issue. I just modify a 900 Maxtrac which has the right ones to start with. 900 MASTR 2's. I have one and there are problems getting them to work. I fooled with the synthesizer in one for a while and did manage to program a micro to load the right code. Then, the VCO wouldn't lock above 901 MHz. You still have to change the front end filters which isn't too big a deal. I just decided I had put enough time and labor into it already. I also have an EFJ repeater that I modified but decided that I didn't like it. Especially the exciter. It takes a crystal for the operating frequency as well as a reference input. The manual says it might take a minute for the VCO to lock up. A minute. The PA looked OK. But that receiver has lousy sensitivity. It does, however have diversity as standard which is good. I just didn't like it. The few Spectras I have fooled with actually work pretty well. Just using the receiver, a lot of the troublesome stuff is avoided. I won't need COS, so there is another problem avoided.

[wa6jbd]

All of the issues in using a Spectra receiver at 902 MHz that are described above can be overcome, relatively easily. I've modified several receivers. I add 0.5to 1.0 PF chip caps across each element of the filter to lower the pass frequency. The VCO modification using the resistor works just fine, or one can solder bridge the laser cuts on the microstrip resonator visible through the rectangular window of the VCO. A hex edit of RSS software will allow programming down to 902. There is a pin location on the audio PA used to turn the audio PA on when receiving a signal that can be buffered and brought out for COR. When completed, receiver performance meets specification on the 902 MHz end of the band.

[John G]

Cool! That is just the kind of information I was looking for. Thanks!

[motorola_otaku]

I add 0.5to 1.0 PF chip caps across each element of the filter to lower the pass frequency.

Have you swept one of these on a spec-an/tracking generator to see what they look like? I'm interested in trying this on a 450-482 UHF radio for the guys who have no need for 470-482 but would still like rated performance on GMRS and a little extra RX on 440-450.

[wa6jbd]

Have you swept one of these on a spec-an/tracking generator to see what they look like? I'm interested in trying this on a 450-482 UHF radio for the guys who have no need for 470-482 but would still like rated performance on GMRS and a little extra RX on 440-450.

Yes, I have. I use 2.0 pf for the UHF radios, and it moves the passband down to about 430-460. You could use a slightly smaller value and not move it down quite as far. As far as performance, filter loss and ripple is essentially identical as before tuning, it just slides the whole filter response down. For your purposes, I'd try 1.5 PF. It'll cover the 440-450 section just fine, and probably still go high enough to cover GMRS.

For those without sweep gear with a desire to just move the filters down to cover the entire ham band, use 2.0 PF chip caps on UHF Spectras. For the 900 radios 1.0 pf works fine to get full receive performance across the band.

I keep thinking I'll write it up and post it somewhere. The same filter mods work to move UHF range 3 radios to the ham band, 900 radios down so they receive at 902, and 800 Spectras that will receive from 806-824.

[wa6jbd]

I know that there is continued interest on this topic. I've received several e-mails from people requesting that I write this up. Being rather busy with a few dozen projects at a time, this hasn't been on the top of my priority list, but tonight I modified a 900 MHz filter for 902-928 operation. I documented the process with close up photos and a few screen shots off the network analyzer. I now have the raw data needed to put together a half decent write-up on the process. I do want to make a correction to my previous post above. On the 900 MHz filters, 1.0 PF centered the response on the low end of the band. A bit of experimenting shows that for a 902-928 passband, 0.5 PF is a more optimum value. The process I will write up requires no sweep gear, just a steady hand, and a magnifying glass.

Although the process is the same to move range 3 UHF filters down to 440 or lower, and 850-870 filters down to 806-826, they require different values. The bench is currently configured for Spectra filter tuning, so I'll do a few examples of the UHF and 800 filters and provide suggested part values. If someone is interested in hosting a PDF or HTML file with the full writeup for Spectra filter tuning, I'm more than a bit motivated to get this info out there.

As for the rest of the conversion process, VCO, programming, etc. I just made use of available info on modifying RSS, and changed the lowest allowable RX frequency to 902 MHz.

edit:

I did some more experimenting, and found that 4.0 PF works pretty good on the UHF filters to center the passband on 430 MHz. I'm seeing about 408-455 at the 3 db points. I also did an additional 900 MHz filter and found that with 0.5 PF, it started to roll off at around 925 MHz. That's good for anyone wanting to use a Spectra on 902 as a repeater receiver, but might be a problem if you're interested in 927-928 RX for mobile use.

More to follow in a few days...

Mel - WA6JBD