2995dx modulation on a scope

[JoeDirt]

I've got a RCI 2995DX in the box too.

What you have done is created a v.cool thread here. I may well do this mod to mine as well now that I have seen what you have done here with yours - 359.

I would just shorten the leads on the resisitor side - if anything. With the waveform you've got as a record for the mod as demonstrated, I am recommending this thread now.

This is the modulator section from a RCI 2995DX and it is the same as every EPT****** style board, this section is around the bottom right corner of the schematic. The difference is that the resistor and capacitor number are different (same values) between the boards. If you can find this modulator section and the resistor I highlighted on any EPT****** board and this mod will work the same on any other radio.

Some people call this mod the PPE (positive peak expansion)

 

[JoeDirt]

If you think that's something, wait til I get my magnifier and I can actually see it :blink:




Yeah, if I leave the trimmer in or replace it with a fixed resistor, I think your right, I'll remove R292, clean up a little with some braid and stick the resistor/diode combo in the R292 hole.

Glad I can help 359

 

[GnG8d]

Glad I can help 359

That you did, many thanks! I've read so many horror stories about sending radios out that I can't even bear the thought.

 

[ExitThirteen]

Now I have a question... after doing the NPC mod, did it change SSB output at all? Seems that when I've put NPC mods into radios, the SSB output drops a bit. I'm totally digging this thread, but I'm curious as to if it changes anything on SSB output. Keep up the great thread!


~Cheers~

 

[GnG8d]

I thought about just buttoning it up,

I couldn't leave well enough alone. I took the 560ohm off of R292 and put in place a 2k trimmer/1n4001 diode series combo in R292. I couldn't get anything but big flat tops. I still had 90/95% negative with 170% or so positive but man were those peaks flat ....... I mean flat. The negative peak looked best at/near zero resistance, and got bigger (less negative) as i added resistance no matter what i did with VR17.

The lighted magnifier is great except I ordered the wrong diopter, I should have gone with 3 instead of 5. 5 doesn't give you enough focal distance, especially when the board is down in the chassis. I'm going to buy a 3 diopter and keep the 5 rather than sending the whole thing back for exchange.

Anyway, I'm going to Radio Shack tomorrow for a small bread board and gather up some different components. I'll just solder leads at R292 and make a small module on the bread board to experiment with. The pads on the board for the resistor are about the same size square as the diode lead is round

For what ever reason the 560ohm resistor in parallel with 235ohm resistor/diode series combo worked! The 560ohm resistor had to be doing some current sharing. I think what I'll do is put 2 trimmers in parallel on the bread board, and put a diode in series with one of them.

Just by chance, the diode wouldn't be the problem would it? I read 1n4001 general purpose/rectifier diode all over the internet for this mod, but it had a 1n4148 switching diode in it.

 

[GnG8d]

Now I have a question... after doing the NPC mod, did it change SSB output at all? Seems that when I've put NPC mods into radios, the SSB output drops a bit. I'm totally digging this thread, but I'm curious as to if it changes anything on SSB output. Keep up the great thread!


~Cheers~

I never noticed much of a power loss before I started messing with it, but i always went from AM to SSB and left the RF power where it was. The tech I bought it from did say that it was doing 225AM, but only 175SSB when I picked it up. He said it got a little dirty above that and the local sidebanders would bitch.

Maybe the mod that I didn't know it had was the reason

 

[Shockwave]

Yes if you set it up for a prefect %98 negative modulation with a 1k tone, the amount of compression does change between different audio frequencies, even at 400hz there is a noticeable change in negative peak compression. I have found that the lower frequencies go to zero volts a lot easier than higher ones. I also have found keeping the factory mod limiter intact helps keep the negative compression more equal across the frequency range of the radio.

The NPC mod is not perfect but in this case with 359's radio it can be improved on. I think for your average radio operator this mod give you the power, swing, and cleanest audio for the little work involved to do it right.

Joe points out some good facts here. Proper setting of the factory limiter should prevent the negative peak diode from conducting very hard and producing sharp corners when it turns on. Two other things play an important role in this NPC circuit. That is the value of the resistor used with the diode and the setting of the carrier. The carrier must be low enough to provide the headroom for the anticipated positive peak without flat topping it. The value of the resistor must be chosen so that it limits the negative peak to about 95%. Too low a value and it will cut the negative peak off too early as 359 saw initially.

It seems like the resistor is being used to control the peak AF voltage at which the diode conducts. If this is the case there may be benefits from using different types of diodes with various forward voltage drops rather then controlling it all trough resistive padding. Typical rectifier diodes such as the 1N4001 do not have ideal switching properties at audio frequencies because they are slow. You may have better results with an LED for two reasons. They react faster and drop more voltage before conducting. I would experiment using an infrared LED first since they drop the lowest voltage of any LED at .9 volts and are most comparable to the original diodes voltage of .6 volts. This could offer improved clamping of the negative peak at 98%.

Joe also notes that the circuit responds differently at bass frequencies then it does with higher AF frequencies. Most broadcast compressors will have more then one setting to adjust the compression independently for the low and the high end response to alleviate this problem. Having said that, I can't argue Joe's point that this mod does offer some nice results when done correctly for very little work. Thanks for providing the scope pictures that convinced me to look at this mod more seriously.

 

[GnG8d]

I thought maybe the 1n4001 could be a little slow, I've got a couple switching diodes here, but I'll grab some more and a couple infrareds too.

 

[JoeDirt]

I couldn't leave well enough alone. I took the 560ohm off of R292 and put in place a 2k trimmer/1n4001 diode series combo in R292. I couldn't get anything but big flat tops. I still had 90/95% negative with 170% or so positive but man were those peaks flat ....... I mean flat. The negative peak looked best at/near zero resistance, and got bigger (less negative) as i added resistance no matter what i did with VR17.

The lighted magnifier is great except I ordered the wrong diopter, I should have gone with 3 instead of 5. 5 doesn't give you enough focal distance, especially when the board is down in the chassis. I'm going to buy a 3 diopter and keep the 5 rather than sending the whole thing back for exchange.

Anyway, I'm going to Radio Shack tomorrow for a small bread board and gather up some different components. I'll just solder leads at R292 and make a small module on the bread board to experiment with. The pads on the board for the resistor are about the same size square as the diode lead is round

For what ever reason the 560ohm resistor in parallel with 235ohm resistor/diode series combo worked! The 560ohm resistor had to be doing some current sharing. I think what I'll do is put 2 trimmers in parallel on the bread board, and put a diode in series with one of them.

Just by chance, the diode wouldn't be the problem would it? I read 1n4001 general purpose/rectifier diode all over the internet for this mod, but it had a 1n4148 switching diode in it.

You sure you had everything installed right? I bet you had too much gain from your tone generator, making your peaks flat top.

I think the separate board is a good idea!

 

[JoeDirt]

You got my interest with the led Shockwave!

I loaned out my galaxy maybe I need to get it back and try that out!

 

[GnG8d]

You sure you had everything installed right? I bet you had too much gain from your tone generator, making your peaks flat top.

Yup, checked and rechecked. All of my gains are identical, in fact, I never touched anything from one test to the other except the board. The flat tops wouldn't budge no matter what I adjusted. They wouldn't go up/down, would round off on the edges, nothing, it's like they were glued there.

I think the separate board is a good idea!

Yeah man, that's some delicate stuff. I never realized how delicate until I got R292 off and looked at it on the end of my finger, it's only 2mm (.078") long I think .......... I've ran engines that took a plug cap that wide

 

[JoeDirt]

The flat tops wouldn't budge no matter what I adjusted. They wouldn't go up/down, would round off on the edges, nothing, it's like they were glued there.

If you turn down the tone generator gain all the way, you shouldn't show any modulation at all? If you were still showing modulation you have a wire crossed or some RFI feed back going on.

Also if your changing the resistance of R292 it does have a affect on the audio drive of the radio. If you making the radio easier to modulate then you could have too much audio gain from your tone generator.

 

[GnG8d]

If you turn down the tone generator gain all the way, you shouldn't show any modulation at all? If you were still showing modulation you have a wire crossed or some RFI feed back going on.

The trimmer and diode were soldered together, and their leads were soldered right to the board, so I'm not sure what was going on. I never changed the generator gain, that was the only constant at 28mv.

ANYWAY, I lifted the trace from the board tonight

I build the circuit that looked the most promising to start with on a Radio Shack generic PC board. With 2 trimmers and a switching diode I had the diode/250ohm in parallel with 550ohm. I was right back to 160%+/95%-, 50w peaking 225. I was attempting to change resistances of the parallel trimmers while noting the change, and I moved one of the wires hard enough to pull it from the board bringing the trace with it.

As an afterthought I realized that I should have added a switch to the 550ohm parallel trimmer to look at a greater number of possibilities, but I'll go back and add that while I'm deciding on how to repair the board

 

[JoeDirt]

The trimmer and diode were soldered together, and their leads were soldered right to the board, so I'm not sure what was going on. I never changed the generator gain, that was the only constant at 28mv.

ANYWAY, I lifted the trace from the board tonight

I build the circuit that looked the most promising to start with on a Radio Shack generic PC board. With 2 trimmers and a switching diode I had the diode/250ohm in parallel with 550ohm. I was right back to 160%+/95%-, 50w peaking 225. I was attempting to change resistances of the parallel trimmers while noting the change, and I moved one of the wires hard enough to pull it from the board bringing the trace with it.

As an afterthought I realized that I should have added a switch to the 550ohm parallel trimmer to look at a greater number of possibilities, but I'll go back and add that while I'm deciding on how to repair the board

I'm saying you should have changed the gain! R292 affects the negatives, and the positives slightly!

You only need one trimmer!

Follow the trace and soldier your wires at the next junction!

 

[GnG8d]

I'm saying you should have changed the gain! R292 affects the negatives, and the positives slightly!

You only need one trimmer!

Follow the trace and soldier your wires at the next junction!

Let me back up here a bit. All I did was put the generator in place of my mic in the audio chain. I unplugged the XLR and plugged in the RCA. So the generator is actually sitting at 1.5mv. From there I have input/output gain control on the preamp/effects processor, and also mic gain control on the radio. So I have several ways to control final input gain ......... I think that will work. I'm using Audacity as a generator and each time you want to change amplitude, you have to create a new file.

Anyway, I'm on a bastard maintenance shift at the plant of 3:30p-1:30a and instead of leaving well enough alone I had to experiment further and lifted the solder pad. I'm going to wait on the 3 diopter lens before I go any farther because the 5 just doesn't give me room to work. I have to see what I want to do and then move the magnifier out of the way and do it blind. Kind of like the arc welding days before auto welding lenses

Instead of going to the cap/transistor that are next in the circuit, I'd like to fix the solder pad @R292 with pins/connector, but I'm headed out the door in a few so I'll see what I can find tonight.

Yup, I should have left well enough alone, but it's that curiosity thing I guess