# LTR Data Slicer Questions

Status
Not open for further replies.

#### rescue161

##### KE4FHH
Can anyone tell me if the components are critical in value on this data slicer? I couldn't find, nor do I believe that the values on some of the resistors are even available.

Thanks.

#### cg

##### Member
While I cannot say if the values are critical, I ordered and received them all from Mouser.com (and it worked when done). Just enter in this part number and change the values for each item
71-RN55D-F-4.42K

chris

#### rescue161

##### KE4FHH
I stand corrected then. My only choice here today was Rat Shack, so I went up to the next closest value and will try it that way.

I'll post my results when it's complete.

#### SCPD

##### Member
Hi Scott;

Pro92b made some comments on that design. The RC values in the low pass filter were calculated for 300 hz. Pro92b suggests 150 hz as the cut-off. I suspect anything in this range will work so you've got some room. If you have to round to the nearest standard value, I suggest rounding up (eg. 4.42k becomes 4.7k).

-rick

#### rescue161

##### KE4FHH
That's what I did. I'm also looking on Mouser now, but for some reason, they're wanting me order a minimum of 4000 on some of the caps. Don't need THAT many...lol

#### rescue161

##### KE4FHH
Well, I finally got it to let me order in small quantities.

Parts are on the way for two.

#### slicerwizard

##### Member
These designs usually call for 1% resistors. If I were building this, I would aim for the correct values. Deviations might affect linear phase. With a good DMM to check values, they're easy to get:

3.92k = (just use a 3.9k, well within tolerance)

4.42k = 4.2k + 220 in series, 39k + 5k in parallel.

11k = 10k + 1k in series, 22k + 22k in parallel.

#### rescue161

##### KE4FHH
I knew you could do that, but space is sort of an issue and I wanted to make a circuit board layout that would be easy for everyone to follow; basically the same thing I did for the Spectra Securenet repeater module.

#### slicerwizard

##### Member
You could wrap the leads of one resistor around the other before installing on the board. Paralleling two 1/8w resistors doesn't use much room.

#### inigo88

Hey guys, I'm a complete novice when it comes to building and testing circuits. I built this slicer on a breadboard just to see if it would work (by no means as a permanent solution), and I can't get it to register more than several hundred interrupts per second on Slicer.exe. As I said, I'm really at a loss on what to change as this is the first circuit diagram based project I've ever undertaken, but I took some pictures. If anyone could take a look and make any recommendations to me if things are blatantly obviously wrong, I'd really appreciate it! Two of the Capacitors are kind of large because they were the only ones my electronics supply store had in stock, but the guy there assured me they were within the limits set on the diagram (although that doesn't necessarily mean much).

I have a feeling I made an error somewhere in the "Comparator" or "Power Supply" sections, as I don't really understand how these sections work. I also could have incorrectly grounded the circuit perhaps? Again, any help would be greatly appreciated... I really want to get this thing running so I can start picking apart my local LTR Passport systems!

Here are the photos, additionally one more that didn't fit can be viewed at http://i17.photobucket.com/albums/b79/inigo88/LTR Dataslicer/IMG_1293.jpg.

Thanks!

-Inigo

#### Attachments

• 167 KB Views: 953
• 177.2 KB Views: 1,089

#### slicerwizard

##### Member
Well, for starters, you haven't indicated what you're using as a signal source; to get any meaningful results out of SLICER, you need to be monitoring a channel that is always carrying low speed data. A very busy LTR or Passport system would work, as would a very busy Motorola analog trunking system. If either of these are available to you, program your receiver to scan all voice channels or talkgroups with no hold delays; that will give you a steady stream of 300 bps data to work with. You can program all the voice channels as conventional channels, or if your tapped scanner can trunk, you can set it up as a trunked system.

An alternate signal source would be an FRS radio programmed with PL tone #24 (151.4 Hz); that will give you a 300 bps datastream of 0101010101...

Run SLICER and press F1 (x1 scale display mode); if you are using an FRS radio as described above, you should see about 302 interrupts per second and when the count hits 8k, you should see one or two wide spikes, or perhaps two overlapping ones. These will tend to be on the left half of the screen. This variability is caused by the differing characteristics of receivers and low pass filters.

Modulating the signal (talking into the mic) should have very little effect on the IPS count; if it does, your slicer is not attenuating higher (voice) frequencies very well. If the IPS count goes up, try talking a bit quieter - the low pass filter doesn't completely attenuate voice frequencies. Obviously, keep your radio away from all those long wires you've used (stand across the room or send a helper outside)

If you go the LTR/Passport/Moto route, you should see a much lower IPS count while monitoring active voice channels, since the datastream contains various runs of matching bits (101111001000011...) which means fewer state changes; 50 to 100 IPS is probably in the ballpark. I get 70 on Moto voice channels. If your numbers are much higher, your filter isn't filtering or you're tuned to a control channel or to nothing (no RF signal)

When the interrupt count hits 8k, you probably won't see much of interest on the display, since most of the useful information will fall off the left side of the screen.

If things aren't working out for you, start with the basics - measure the voltage on the IC's power pins (while you are running SLICER); you should see at least +/- 6V; +/- 9 or 11V would be better. Measure all the resistor values; you can do it in-circuit with a digital multimeter if the power is off (disconnect from the PC's serial port)

That should be enough to keep you busy for a while...

#### rescue161

##### KE4FHH
A few of those look like polarized capacitors to me. Not sure if that would make a difference.

#### inigo88

Rescue161: I wasn't sure either about using several polarized capacitors, but they were all I could get at the time. I head that if you face the (-) side of the capacitor towards ground you're in good shape, but I could have been misinformed. Thank you for your insight.

I used both side rails of the breadboard as ground and tied them together with a wire to make it a common ground, but I honestly have no idea whether that was a logical thing to do or whether it's adversely affecting the whole thing. Also, I read somewhere online that it was a good idea to join the wire that comes from the GND pin of the DB-9 connector with the bare copper stranded ground wire that comes from the connector housing itself, and send them both to the ground rail on the circuit. Same question, was this the logical thing to do or completely incorrect?

SlicerWizard: THANK YOU VERY MUCH!!! I don't own an FRS Radio or HAM H/T so I can't generate the PL tone, but I do live within close proximity of several busy LTR Passport and Standard systems along with several Motorola systems (the busiest and strongest being a digital Motorola 3600 bps Smartzone system which I'm not sure the subaudable data would work the same way on or not, but could at least try). The Passports may also be excellent candidates because they have the idle burst every 3 to 5 seconds even when there is little to no voice traffic. I DO own a digital multimeter and will get to work and try both measuring the voltage while it's plugged in and verifying the correct resistance of all the resistors when its not.

One final question for you - Since the diagram doesn't specify an RS-232 output pin, do you prefer one in particular? I've heard DSR and CTS will do the job, but do you know does any one have the best results with slicer.exe and LTRDump/LTRTrunk?

I can't thank you guys enough. I'm 18 years old and have only been into scanners for a year or so now, but I tried to hit the ground running so to speak and learn everything I possibly can about trunking protocols and systems. I've extensively examined Motorola and EDACS systems in my area and got intrigued by figuring LTR Standard out and bought a scanner that could track it as well. I recently mapped out and identified the users of two local LTR Standard systems and submitted them to RR.Com, and there seems to be an abundance of LTR Passport systems in the area that I'd love to figure out and identify the users on.

Thanks again for helping me get back on track!

-Inigo

#### slicerwizard

##### Member
I wasn't sure either about using several polarized capacitors, but they were all I could get at the time.
The caps specified in the schematic (in the power supply section) are designed to absorb noise spikes that are generated by high speed switching. The electrolytics you're using don't do that as well, but probably won't cause any problems. I see you've also used one in the comparator section; I think it'll be OK, but do yourself a favour and get the correct ones down the road.

I head that if you face the (-) side of the capacitor towards ground you're in good shape, but I could have been misinformed.
The negative side goes to the more negative point in the circuit. In the schematic's power section, where the two 1.0 uF caps are shown, your substituted electrolytics would have both of their (-) markings at the bottom, so your lower one is probably in backwards right now. It's a good thing serial ports can't pump out a lot of current; if they did, that cap would've blown up. No, I'm not kidding.

I used both side rails of the breadboard as ground and tied them together with a wire to make it a common ground, but I honestly have no idea whether that was a logical thing to do or whether it's adversely affecting the whole thing.
That's the way I did mine last night. Nothing wrong with that.

Also, I read somewhere online that it was a good idea to join the wire that comes from the GND pin of the DB-9 connector with the bare copper stranded ground wire that comes from the connector housing itself, and send them both to the ground rail on the circuit. Same question, was this the logical thing to do or completely incorrect?
It's not necessary and won't have much effect here.

I don't own an FRS Radio or HAM H/T so I can't generate the PL tone,

but I do live within close proximity of several busy LTR Passport and Standard systems along with several Motorola systems (the busiest and strongest being a digital Motorola 3600 bps Smartzone system which I'm not sure the subaudable data would work the same way on or not, but could at least try).
300 bps data is 300 bps data. SLICER doesn't care.

The Passports may also be excellent candidates because they have the idle burst every 3 to 5 seconds even when there is little to no voice traffic.
Those bursts are so short that they won't help you diagnose anything. You need voice comms.

I DO own a digital multimeter and will get to work and try both measuring the voltage while it's plugged in and verifying the correct resistance of all the resistors when its not.
Cool. If you have them, clip leads are a good way to get accurate readings.

One final question for you - Since the diagram doesn't specify an RS-232 output pin, do you prefer one in particular? I've heard DSR and CTS will do the job, but do you know does any one have the best results with slicer.exe and LTRDump/LTRTrunk?
CD, DSR and CTS are all suitable output pins. They are all treated the same by SLICER. Most decoding applications should also work with any of the three pins. My personal preference is CTS (pin 8)

#### inigo88

The negative side goes to the more negative point in the circuit. In the schematic's power section, where the two 1.0 uF caps are shown, your substituted electrolytics would have both of their (-) markings at the bottom, so your lower one is probably in backwards right now. It's a good thing serial ports can't pump out a lot of current; if they did, that cap would've blown up. No, I'm not kidding.
Understood, and thank you for setting me straight! Luckily all three 1.0uF capacitors are non-polarized and marked "NP" in the picture. If you think it would help, I could flip it anyway. The only polarized capacitor I believe is the .033 uF one.

Thanks again,

-Inigo

##### Member
What about a greenboard and Casing and layout to use? to make it clean built and somewhat durable after its built ?? and really i dont understand how to read schematics and follow them to build somthing,so uhhh what should i do ?

Last edited:

-Inigo

#### inigo88

Hey guys, I hooked up my RS-232 data to the CTS pin of the DB-9 connector, measured 8V flowing through the power pins of the IC, and checked the resistance of my resistors. I'm definitely getting a much more dynamic reaction from slicer.exe, and I took some screenshots to show you below, under a variety of conditions.

This is with the slicer mode switch in MOT/EDACS mode (bypassing the filter section) while receiving IMBE digital voice traffic from this site: http://www.radioreference.com/modules.php?name=RR&sid=1316

This is the slicer again in MOT/EDACS mode. I believe it's also recieving voice traffic but even when the scanner was quiet the interrupt rate appeared to remain fairly constant (right around 200).

Here's the slicer in LTR mode (filtered).

In LTR mode with no audio, the interrupt rate shot WAY up.

In LTR mode with voice traffic from the motorola system, the interrupt rate came down and settled during the voice traffic, then shot back up to around 1500 bps. Every time there was voice traffic though it came down to 200-400 IPS.

Here is LTR mode recieving voice traffic with the Invert switch in the opposite position as before.

I don't really know what to make of this, it still doesn't work, but at least it's dynamically reacting rather than sitting parked on one set interrupt rate like it did before when it REALLY didn't work. I tried to cover all the variables by moving the 2 switches around but any further questions about how it behaves in a certain configuration let me know. I'm really encouraged guys! It's showing a lot of improvement and I'd really like to get this thing to work!

Thanks again for the excellent, excellent help.

-Inigo

Last edited: