Modding PLL UPD858 in Midland 77-883

[Aztec]

Hey everyone

I'm trying to modify a new radio. It's a Midland 77-883 AM only radio. PCB is marked as PC-152AC. It has an 858 PLL which I was really looking forward to modding until I ran into a few issues. Perhaps they're not really issues as much as confusion on my part.

Here is the website where I'm getting the mod info from.

http://kg4zpn.netai.net/cb/pllfreqmods.htm

Specifically the very first mod under the list of UPD858 mods.

The problem is that the circuit in the radio doesn't quite match what's in the diagram shown here. In the diagram pins 20,22, and 23 are shown as grounded. Pin 21 is shown as connected to 5V. Pin 19 is where you cut a trace and add a resistor with the switch.

The configuracion in the radio is completley different however. Pins 20 and 21 are connected together. 23 is grounded and 22 I believe is used for binary input.

The mod info does not mention anything about grounding the pins the way it shows them in the diagram so I think it assumes that the radio comes this way from the factory. The only pin that I could really manipulate according to the diagram is pin 19. The rest I wasn't sure what to do with. I was wondering if I could make the rest of the pins match what's on the mod diagram but I'm afraid of damaging something. With pin 19 set up as the mod diagram states but the rest of the pins left untouched, the mod does absolutely nothing.

Does anyone know what can be done about this situation? Can the radio be modded at all? The radio uses a 10.240 crystal. I can't find any other info on this radio or on this chassis.

Attached is a picture of the chassis. The PLL/VCO circuit is in a completley seperate module that solders onto the board.

Your help is much appreciated, thanks!

 

[nomadradio]

The binary arithmetic on this radio is genuinely odd. The uPD858 is technically a 399-channel chip. The radio uses only 40 of those 399 possible binary codes. Should be easy, right?

But the math on this one makes a mod tricky. Pin 22 is connected both to a wire from the channel selector and to input pins 1 and 2 of IC2, a CD4011. This chip inverts the logic level from pin 22, and puts the binary opposite of that onto pins 20 and 21. They are both tied to the output of this section of IC2 pin 3.

To top it off, the chip's inputs are not straight binary, but binary-coded decimal, or "BCD". As a result, not every combination of high and low on the input pins will work like the binary-input chips.

Input pins 13,14,15 and 16 get you a value of 1, 2, 4 and 8, just like normal binary. But pin 17 has a value if "10" when it's high, and zero when low. Pin 18 counts up 20, pin 19 is 40, pin 20 is 80, pin 21 is 100, and pin 22 is 200.

Sams CB161 has a binary truth chart for the 858 chip's inputs. Later volumes would also show the frequency feeding into pin 11 of the chip. This made figuring out the math a lot easier. The Sams data for this radio leaves out that detail from the truth-chart page. Just have to do the binary math the hard way for this model.

For channel 1, all the channel-selector outputs to the 858's input pins are zero, but pin 14. But this is not binary "two" the way it looks.

Pin 19 is zero, but pins 20 and 21 will be at a logic high state. The BCD math for this gives us a divisor number to the chip of 182. Channel 2 is 183, and on to channel 15, at a divisor number of 199. Now, we don't go directly to 200, since that would be channel 15A, a RC channel that the radio must legally skip, going to channel 16.

On channel 16, pin 22 changes from low to high. This causes pins 21 (100) and 20 (80) to go low. Pin 22 being high now starts out our arithmetic at 200. Pins 21 (100), 20 (80) and 19 (40) are all now logic zero. The first pin (after pin 22) thats high for channel 16 is pin 13. This gives us a divisor number for channel 16 of 201. The divisor for channel 40 is 226.

Since pin 19 has a value of 40, and is soldered to ground (logic zero), you would think that simply cutting the ground trace, and connecting a toggle switch that selects logic high (5 Volts) or low (ground) would bump it up 40 channels, adding 40 to the chip's divisor input.

Sure enough, this works from channel 16 and up, but not on channels 1 to 15.

The mods shown in the link above are for later radios that had the 858 chip mounted on the main pc board. This radio, with the metal-box PLL uses different binary arithmetic.

To reduce the risk of damaging the input circuits in the chip on pin 19, I recommend first soldering the 4.7k 1/4W resistor from pin 19 to a spot on the ground foil. You'll probably need to scrape the mask paint from a spot on the ground foil for this. Then, only after the resistor is in place, cut the foil trace that connects pin 19 to ground. Placing the resistor so it won't block this part of the foil is a good idea. This way, the pin is never left disconnected from the radio. Tiny surge voltages on the input pins of this chip can blow it out. The resistor maintains a connection to ground while you're cutting that trace.

A SPST switch from 5 Volts (like pin 12) to pin 19 will force pin 19 high when closed, and the resistor returns the pin to ground when the switch is open.

Problem is, this only works from channel 16 and up. The reason is pin 20. When that one is low, pin 19 works like you expect. But when pin 20 is high, on channels 1-15, pin 40 doesn't work like you expect.

Oops.

You should get channels 55 to 80 on channels 16 to 40. Getting channels between 40 and 55 requires more hookup that I just don't remember at the moment. I'll poke about and see if I can find some old notes on that detail.

73

 

[LeapFrog]

I think Unit_399 has created a truth chart for this PLL (Thanks, Brother).
Take a quick read over of this recent thread here, it may contain some useful information for what you are trying to do.

Some info floating around online is meant for the 858 Chassis radios, not just any radio with a 858 PLL.

The Cobra 21X and Realistic TRC-452 use the same style of circuit board/pll can layout.
& Comparing the 21XLR to the 452 shows they are almost the same exact radio, then comparing the schematics between all three models I found some interesting things out about pins 19,20,21,22 being tied together and or to ground.

Your Midland was made by Uniden and will likely have a brother/sister radio using the same pcb with another mfg.'s name on it. Finding mod info for one model name/ variation may yield better results compared to just searching by pll/ board #'s. But the 858 PLL info can be applied to almost any radio with the 858 you just have to pay attention to the pins/grounding & "PLL can" layout.

Best Regards.
-LeapFrog

 

[unit_399]

On this rig, the "N" number ranges from 182 (CH1) to 226 (CH40) for the std 40 channels. Putting pin 19 high adds 40 to the count. Adding 40 at channel 5 (N=187) on the selector gives us N = 227 (CH41), and the radio will go up to CH54 from there. Problem is ... this is an AM-only rig, and most everything above ch 40 is ssb-land. The radio can be modded to go below chan 1 ( where most of the freeband AM action is) by disconnecting pin 20 from pin 21, cutting traces, and installing some isolation resistors as shown in figure 1. Wire up a DPDT - CENTER OFF switch and connect the wires directly to the pins as shown in Fig. 2. Putting the switch in the "low" position will give you 26.955 at selector position 40 and the rig will go down from there. With the switch in the 'high' position you'll get 41 thru 54 . In the center-off position you will have the std 40 channels. Good luck & 73s.

- 399

 

[Aztec]

Thanks everyone for all the information and sorry for the late reply, I had gone on vacation for the holidays, speaking of which I hope you had some great holidays and I wish you a happy new year!

Unit 399 thanks a lot for that mod info. Looks like it's just what I need. I will absorb the rest of the info you guys have to offer regarding the 858 PLL. I really need to polish up on my understanding of these chips. Though I more or less know the basics of PLLs, the specifics I'm a bit rusty about. In the future I hope to experiment with some microcontroller mods or possibly some digital control VFOs.

 

[Aztec]

Unit 399,

I finally got around to performing the mod but I'm stuck at the switch wiring. I'm a bit confused by the diagram where it says to hook one throw of the switch up to 18, 19, and 21. Am I supposed to connect all of those three pins together together? If so what would be the point of having the other side of the switch connected to only 19?

 

[unit_399]

After installing the resistors, and making the trace cuts, if you take apply 4.95v to pin 19, the radio will operate on chans 41-54. Applying 4.95v to pins 18, 19, 21, while the other contact set in the switch applies ground to pin 22, the radio will operate 40 chans below the regular cb freqs. With the switch in the center (off) positions you have the regular 40.

- 399

 

[Aztec]

Ah okay understood.

I'm guessing in that case I'll have to isolate pin 19 so when the switch is up, only pin 19 gets 5V. I'm thinking a couple diodes that come from pins 18 and 21 so that when pins 18 and 21 get sent high, the voltage will go to 19 from them. On the other hand when the switch is high the switch will send voltage directly to 19 and the diode will block it from getting to 18 and 21.

Would that work?


EDIT: This diode idea might affect the radio while it's in "normal" operation if either 18 or 21 go high so will 19. Other than that though I am not sure how I could only energize pin 19 on one side of the switch if on the other side of the switch 18, 19, and 21 are all tied together.

 

[unit_399]

I am not sure how I could only energize pin 19 on one side of the switch if on the other side of the switch 18, 19, and 21 are all tied together.

I should have noticed that. Connect the switch to the 858 chip as shown below. Solder the switch wires directly to the 858 pin solder pads. Use 1N194 diodes or equiv. Observe polarity. Sorry ... my bad.

- 399

 

[Aztec]

that's perfect! I'll report back when the mod is done.

 

[Aztec]

Okay I did it and it works though not how it was predicted.

The mod operates on the channel selector from channel 16 to 40.

On high 16-40 gets me from 27.555 to 27.805.

On low it gets me from 26.755 to 26.935 on channel 33. After channel 33 it goes back down to 27.745 on channel 34. From there it ends up at 27.805 on channel 40.

Little bit weird. As far as I can tell I wired everything correctly. Need to go over my work one more time. Would be nice to have all 40 channels on the lower frequencies. If not then I'm happy with what I got but since you seem to be talking about this same PLL module I'm wondering why I'm getting a different result.

 

[Aztec]

Okay well I've been trying to figure out how I could make this thing work so that the entire dial becomes lower frequencies that are continous when this thing is placed on low.

What I found suggests that this mod as is, may not actually be able to start at 26.955 on channel 40 and go from there.

I will try my best to explain my findings. But basically the mod cannot perform some needed changes in the BCD from channel 1 to 15 and from channel 33 to 40. The main issues is the behaviours of P7 (Pin 20), P6 (Pin 19), and P5 (Pin 18). One thing I find peculiar is that PIN 20 still switches high and low even when the mod switch grounds pin 22. It seems the nand gate (IC3) does not detect us grounding pin 22 via the mod switch and only responds to the change in the channel selector.

I made a truth chart for this by basing myself on channel 1 being 182 and channel 40 being 226 and taking into account all the RC channels by comparing the N number to a CB frequency chart. In the end I got a channel vs N code chart that matches. If you guys want it let me know so I can slap it on an excel file or something.

But basically my lower channels only work from 16 to 33. On channels 34-40 and 15-1 I get stuff that's out of place.

NOTE: I wrote my BCD codes backwards with the lowest value digit on the left side. I dont think this is the correct format, I did this because that's how another truth chart that I found was formated. I will correct it if I have to.

If for example we wanted 26.955 the N number for that would have to be 181. The first place I can manage to make 181 using the BCD code is on channel 35. Channel 35 is normally 1000010001 which is N code 221.

Now, the mod that we did, when on the low position gives us 1000011010 on channel 35 which is N code 161 which is already on channel 16 and gives us 26.755.

In order to turn this into 181 while the switch is in the low position we must do the following:

P7 aka Pin 20 must be made high. Since the mod switch is grounding pin 22 I would expect pin 20 to already be high due to IC3 the nand gate but it isnt.

P6 aka pin 19 must also be made low. This is being held high by the mod switch.

P5 aka pin 18 must also be made low. This is being held high by the channel selector itself on channels 34 to 40 and also the mod switch.

With those changes to P7 P6 and P5 aka pins 20, 19 and 18, I should get non repeated lower channels from 27.005 on channel 40 to 26.945 on channel 34. Keep in mind that once the channel selector goes to 33 the original mod, as prescribed by you Unit_399, begins to work starting with 26.935 on channel 33 all the way down to 26.755 on channel 16. Everything is merry until we get to channel 15 were everything gets screwed up again.

Now channels 1-15. I need to recheck exactly what the deal is with this part of the coding as it's stands now but here's how it can be made to work continously all the way to channel 1.

Pin 20 P7 must be made low, I believe it is being held high by the nand gate.

Pin 19 P6 must be held high as it already should be by the mod switch

Pin 18 P5 should be made low, currently it is held high by the mod witch.

This should give 26.735 on channel 15 aka N code 159 BCD 1001101010 and onward until 26.585 on channel 1 N code 142 BCD 0100001010 if I got that right.


Alright well I was practically falling asleep by the end of making this post. I hope I didn't make some major mistake in my thinking and I hope you guys get what I am trying to communicate. Please excuse any mistakes I made have made in there! Tell me what you think.

 

[Harley3315]

Anyone have copy of unit_399's web page it was like 20 some pages of awesome info on the 858 chassis radios the site is gone

 

[Klondike Mike]

Anyone have copy of unit_399's web page it was like 20 some pages of awesome info on the 858 chassis radios the site is gone

See attached file.

 

[500kpot]

Hello Astec

Here is a metric butt load of information on the 858. Some may apply and some not but I sincerely hope it does help you.

Page 4
http://cbradiomagazine.com/CB Secrets Web/SECRET CB VOLUME 12.htm

Page 44
http://cbradiomagazine.com/CB Secrets Web/SECRET CB VOLUME 5.htm

Page 56
http://cbradiomagazine.com/CB Secrets Web/SECRET CB VOLUME 7.htm

Page 8 <- This is very informative
http://cbradiomagazine.com/CB Secrets Web/SECRET CB VOLUME 9.htm

Page 4 & 35
http://cbradiomagazine.com/CB Secrets Web/SECRET CB VOLUME 11.htm