Question about ceramic resonator again.

OK this is really my first time dealing with these things and my Google searches are not turning up what I'm looking for. So my observations are these and Mike or anyone, see if this sounds right to you because as I said I am not getting the answers I want off the internet. It looks as if they are highly prone to harmonics meaning in this case, that if the resonator is 455KHz, it will also allow 910Hz and 1820KHz and 3.640MHz which seems to be dominant and on up. But the frequency around 3MHz seems to be prevalent and one or two multiple of that. It first I thought I was getting intermod so I changed my mixer to a double balanced passive mixer and reduced the gain of the front end. I see now why certain frequencies reoccur. It is not intermod it is the harmonics. So my question is this. Is it common practice or taboo to use a ceramic resonator as a 1st IF filter? Maybe there are better ones then the cheap one I bought but I am very dissatisfied with it's performance since my intention was to use it in a single conversion receiver design. In fact I even used two of them. One after the post mixer amp then a single stage amplifier followed by another. It helped but I still had the same harmonic responses.

The use of 455 kHz I.F. depends on the RF frequency. Usually you want RF to IF ratio of 5 to 7. Too high RF will require too much RF filtering to reduce mixer image. Too high an I.F. will be subject to A-B (Able-Baker) spurs.

You can always go to double or triple conversion.

Most new receivers are direct conversion. The mixer is very precise to avoid D.C. offsets and 2nd order crossmodulation. There is usually a automated self calibration within the direct conversion I.C..

RCinFLA said:

The use of 455 kHz I.F. depends on the RF frequency. Usually you want RF to IF ratio of 5 to 7. Too high RF will require too much RF filtering to reduce mixer image. Too high an I.F. will be subject to A-B (Able-Baker) spurs.

You can always go to double or triple conversion.

Most new receivers are direct conversion. The mixer is very precise to avoid D.C. offsets and 2nd order crossmodulation. There is usually a automated self calibration within the direct conversion I.C..

Click to expand...

Wow! Let me digest this a minute. Sounds like you have some experience. So what sort of mixer do you recommend? A 1496?

edit* Automated self calibration. I got to look that one up. So you say most are single conversion? I am surprised indeed.

edit edit* Is this with PLL systems only? If I follow this correctly, you are talking about the output level of the VFO or VCO? It makes sense. I have noticed that re-tapping the coil will change the performance to a degree but I still think the ceramic filter is more sensitive to harmonics and that it could be cured through a crystal filter. So if this is PLL only, than what would be the cure in a low end VFO type receiver?

Maybe your over driving the filter input causing input/output isolation issues. I would not go over 0 dBm at the filter. You can cascade the filters for better selectivity but the murata data sheets suggest that you put a buffer in between.

Mikebits said:

Maybe your over driving the filter input causing input/output isolation issues. I would not go over 0 dBm at the filter. You can cascade the filters for better selectivity but the murata data sheets suggest that you put a buffer in between.

Click to expand...

I got the output buffered. I just don't know if it is common practice to use one of these ceramic filters for a single conversion where it is the only IF filter. It just seems kind of unreliable. I see them all the time in 2nd IFs and it would be useful for that but the real question is using that as the whole receiver's selectivity. You now what I'm saying? Looking at the internet I have not been able to pull up any schematics where a 455KHz ceramic resonator was the only filter.

Maybe you can build a stand alone filter eval board and test it's characteristics with known in/out impedance and drive levels. I am sure you have a sig gen you can use no? By testing it in stand alone you will get an idea of how it will perform in a circuit. When I worked in a RF lab we always evaluated parts this way prior to putting it in a circuit.

Are you using a ceramic resonator, or a ceramic filter? There's a significant difference. A resonator is intended to be used in oscillators, and usually has only two terminals. A ceramic filter, as the name implies, is intended for use in filters and will have 3 terminals.

Ceramic filters are eliptical design meaning they have significant frequency response flyback after the close in null. Ulimate rejection is about -30 db for two pole versions. Having two filters separated by an IF amp will yield reasonable outband rejection.

You can make a filter from two resonators and a transformer but why do it when self contained filters are available and cheap.

Auto adjustments (by a small processor) for a zero IF direct conversion rcvr includes DC offset centering, 2nd order optimization by gain/phase adjustment on double balanced mixers in the I and Q channels.

2nd order crossmod are two close by strong signals creating low frequency product that would pass through a baseband (low frequency) zero IF (direct conversion) receiver. I and Q channel must have matching gain and phase response for the quadature zero IF I and Q channels processing.

Another receiver type used a lot for GSM phones is very low IF. An IF of about 150 KHz is used with the mixer image placed at zero frequency. It only gives about -35 db image rejection but this is okay since adjacient channels are not assigned active in a given cell coverage area.

In this case, SV is using his radio for HF ham radio I believe. Not sure if most ham operators are using IQ modulation.

BobW said:

Are you using a ceramic resonator, or a ceramic filter? There's a significant difference. A resonator is intended to be used in oscillators, and usually has only two terminals. A ceramic filter, as the name implies, is intended for use in filters and will have 3 terminals.

Click to expand...

It's a filter then. It has 3 pins. Thanks for clearing that up

RCinFLA said:

Ceramic filters are eliptical design meaning they have significant frequency response flyback after the close in null. Ulimate rejection is about -30 db for two pole versions. Having two filters separated by an IF amp will yield reasonable outband rejection.

You can make a filter from two resonators and a transformer but why do it when self contained filters are available and cheap.

Auto adjustments (by a small processor) for a zero IF direct conversion rcvr includes DC offset centering, 2nd order optimization by gain/phase adjustment on double balanced mixers in the I and Q channels.

2nd order crossmod are two close by strong signals creating low frequency product that would pass through a baseband (low frequency) zero IF (direct conversion) receiver. I and Q channel must have matching gain and phase response for the quadature zero IF I and Q channels processing.

Another receiver type used a lot for GSM phones is very low IF. An IF of about 150 KHz is used with the mixer image placed at zero frequency. It only gives about -35 db image rejection but this is okay since adjacient channels are not assigned active in a given cell coverage area.

Click to expand...

Man, I love your explanations. This guy is deep. Anyway can I put it to the group this way. Have any of you known of any manufacturer to use a 455Hz ceramic filter as the selectivity device in a receiver. Now I have a mechanical filter for 455KHz I got out of a Collins I would trust but I don't trust this guy (the filter). I'm thinking if a manufacturer did use one than it is a piece of junk. I bought one with a hand crank and it had a little solar cell and it received the same way as this. You would here the same broadcasts all down the band. Wasn't a fluorescent light or power line it wouldn't pick up though.

There are all different kinds and qualities. The really cheap radios use really cheap parts. But yes, I have a radio or two that use ceramic 455 kHz filters and they work very well. Is there any sort of part number stamped on your filter?

Collins mechanical filters are top of the line and very expensive. Consider yourself lucky to have one, and lets hope it is not broken as it would make a excellent 455 Khz filter.

BobW said:

There are all different kinds and qualities. The really cheap radios use really cheap parts. But yes, I have a radio or two that use ceramic 455 kHz filters and they work very well. Is there any sort of part number stamped on your filter?

Click to expand...

Yeah but is it the only selectivity device the radio uses? You now, single conversion. But you make a good point. These where dirt cheap. I got a bag full of them for about 4 bucks. I tried to get a data sheet and they discontinued them. I think I better scrap this filter.

Yeah Mike, the Collins would be nice but they are sort of bulky. I want this radio to be small which is why I'm doing single conversion. Do the break easy or something? I never got a chance to check it out. Maybe I should try it here but not sure of the hoop-up. It's got 4 pins.

Is it like 3 inch long, 1/4" wide, does it have soldered bottom? If so it is a premium mechanical filter, and can easily be damaged if mishandled but the performance is unsurpassed.

----------------------edit------------------------

Hook up is easy, one input, one out and two ground connections. The distance also gives inp/out isolation from radiation.

Mikebits said:

Is it like 3 inch long, 1/4" wide, does it have soldered bottom? If so it is a premium mechanical filter, and can easily be damaged if mishandled but the performance is unsurpassed.

----------------------edit------------------------

Hook up is easy, one input, one out and two ground connections. The distance also gives inp/out isolation from radiation.

Click to expand...

I think it is plastic or maybe a ceramic material but the dimensions sound right. What happens if you mis-handle them? I've seen drawings of them before. They looked like round discs on on a shaft.

hmm, plastic, not what I was thinking, but yeah, the ones I used were made with small circular disc soldered about 1/16 inch apart. Excessive mechanical handling can jar the tuning of the filter. But not sure how robust the plastic versions are. Still most likely a better choice over your ceramics.
Suggest you make a test rig for it. Got a pn#

Mikebits said:

hmm, plastic, not what I was thinking, but yeah, the ones I used were made with small circular disc soldered about 1/16 inch apart. Excessive mechanical handling can jar the tuning of the filter. But not sure how robust the plastic versions are. Still most likely a better choice over your ceramics.
Suggest you make a test rig for it. Got a pn#

Click to expand...

I'm not sure if it's plastic. I was just looking for it. It might be a ceramic material come to think of it. But I have heard really good stuff about them. The more I think about it, I could use that thing to solve my main question and I might keep it if I can fit it in my little project box.

With today's technology, why not use an active BP filter?

Linear Technology - LT1568 - Very Low Noise, High Frequency Active RC, Filter Building Block

----------------------------edit---------------
On second look, that may not be a good part...

Why not just buy a good quality ceramic filter. Toko and ECS both make good ones. I have a bunch of Toko AHCFM2 455kHz filters and they work just fine. I used one in a 455kHz signal generator to remove the harmonics from a 455kHz square wave. I ended up with a very clean sine wave.
Schematic:
**broken link removed**