Another 2 cents worth. You are right about the Silicon Systems DTMF decoders the 202 and 204 they were not quite telephone quality. they were designed by Silicon Systems (SSI) and made under license by other companies and used SSI's mask also licensed. But the CD22204 was a different mask I think it was made by RCA. It was one of the worst decoders that I ever had the misfortune to run into. It would just quit decoding. The reference made about "(47uF or greater) in the power +5 line other wise on a power surge in the circuit it does this weird stuff like it will stop decoding, Force the stobe pin high, then sit and start to "bake" in the socket." is called latchup. That is what happens to an CMOS IC when you go out of MAXIMUM specs on VCC.
The 8870 is exactly the same part as the pair of ICS that was refered to, the 8860 & 8865 (MT8860 MT8865). The MT8860 MT8865 pair was the first IC made and used by the phone company for DTMF decoding, prior to that the decoding was done with a L C circuit. the company that gave us that nifty IC was MITEL who made PBXs (private branch exchanges) (small telephone switches) for companies. These parts are also licensed by a number of manufacturers. ( Most engineers or companies will not design in a part with a single source.) The MT8880 (family) is the MT8870 with A DTMF encoder included. The encoder is very similar to the MOSTEK 5089 encoder. I know this because I have beel involved with the design of telephone and telecommunications equipment for a number of years. --- In [email protected], Matt Krick <[EMAIL PROTECTED]> wrote: > I have to throw my $0.02, So my research says..... > > I always had a preference for the SSI-204 and SSI-202 but my later > designs use MT8870s for the simple reason that they are still being made > by a variety of manufacturers and the 204's are no longer availible > unless you find a surplus vendor which will have them for 5 to 13 bucks, > They also go by CD22204 and a MC number, MC14513? Can't remember. > > The SSI-204 falses fairly easy. > But on the plus side it decodes about 5 mS faster than the MT8870. Down > side is it is tough to keep it happy. You have to put a stiff capacitor > (47uF or greater) in the power +5 line other wise on a power surge in > the circuit it does this weird stuff like it will stop decoding, Force > the stobe pin high, then sit and start to "bake" in the socket. Wear an > oven mitt when experimenting. > > Using the MT8870 reference design, the fastest reliable tones I could > decode are 37mS long with 17mS of silence in between digits. It will > decode with 32mS on time but it is not reliable on strings, With strings > 32mS and 13mS off, it has problems with low and high tones such as > 1,2,4,#,C,D everything else will decode okay. > > Board space as a factor, the 204 uses a 14 pin dip and needs only a > crystal, oscialtor resistor and optional low pass filter of a resistor > and capacitor (6.8K and 0.01uf). where the 8870 is 18 pins and needs a > few more external components to operate and the optional low pass > filter. The 8870 also has pins so you can implement hardware "guard > time" which is kind of nice as it helps with the falsing, and you can > control the gain of the incoming op-amp which my eleminate an extra chip > from your board.. > > All in all the best DTMF decoder I have ever used is that of the ACC > RC-850, Not the fastest but it is the most immune to noise I have worked > with. It uses 2 chips, I believe they were MT8860 and MT8865, Which are > no longer made. Basicly one chip pre-seperated the tones into 2 groups > and fed it to the other to be coded into Hexadecimal. But you can be S-2 > to S-5 into the repeater and still have reliable decode. Yahoo! Groups Links To visit your group on the web, go to: http://groups.yahoo.com/group/Repeater-Builder/ To unsubscribe from this group, send an email to: [EMAIL PROTECTED] Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
