SBE Sidebander VI PLL problem

[TM86]

Sidebander VI, powers on, shows "00" on the display.

Check the microprocessor signals, it's got no clock input. Trace that back to the PLL board, which is supposed to be putting out a 5 MHz signal, it's not.

Checked the PLL (NDC40013) , it's got power (5.6 V), ground, and 10 MHz in on pin 3. Despite the otherwise excellent Lou Franklin tome "The CB PLL Data Book" showing PLL pin 4 as "NC", it's the source of the 5 MHz signal that's missing. Which tells me the PLL is probably dead. Not like the obvious crude soldering all over it wasn't the first hint.

Quick look around the intertubez shows almost no info on the chip, and none available.

So, just how screwed am I?

 

[TM86]

24 hours, 34 views, no comments.

I'll take that as "totally screwed."

Which means I'm going to have to get creative with this one.

 

[guitar_199]

I have not looked into this radio at all... but perhaps a VFO module could "save it"? If you can find one that will run the right frequency and "scotch it in there".

 

[nomadradio]

Seems to me the chip to blame is the one that feeds binary bits INTO the PLL. Looks to me as if this is where the problem lies. A 28-pin custom chip on the front-panel circuit board drives the channel digits and the PLL's binary input pins. Can't remember seeing the mobile version of this radio, only the base radio.

Just checked the schemo here: https://www.cbtricks.org/radios/sbe/sidebander_vi/graphics/sidebander_vi_main_sch.pdf

The chip that's controlled by the keypad is shown as "MM5799MBR". It's a microprocessor/microcontroller of the late-1970s variety. Totally custom. Program lies in the optical pattern used to build the layers in the chip. The photomask that does this is why this kind of permanent program storage is called "Mask ROM". Tweren't no 'flash' chips on the planet back then.

Wouldn't hurt a thing to make sure the chip has 8 Volts on pin 15. If anything interrupts the power to it, this could explain seeing channel "00" on the display.

73

 

[TM86]

Seems to me the chip to blame is the one that feeds binary bits INTO the PLL. Looks to me as if this is where the problem lies. A 28-pin custom chip on the front-panel circuit board drives the channel digits and the PLL's binary input pins. Can't remember seeing the mobile version of this radio, only the base radio.

Just checked the schemo here: https://www.cbtricks.org/radios/sbe/sidebander_vi/graphics/sidebander_vi_main_sch.pdf

The chip that's controlled by the keypad is shown as "MM5799MBR". It's a microprocessor/microcontroller of the late-1970s variety. Totally custom. Program lies in the optical pattern used to build the layers in the chip. The photomask that does this is why this kind of permanent program storage is called "Mask ROM". Tweren't no 'flash' chips on the planet back then.

Wouldn't hurt a thing to make sure the chip has 8 Volts on pin 15. If anything interrupts the power to it, this could explain seeing channel "00" on the display.

73

IC 802 has 8.49 volts on pin 15. What it doesn't have is a clock signal into pin 7. I traced that back through several other components to arrive at the source, PLL pin 4.

There is a 10 MHz reference clock signal going into the PLL on pin 3. There's currently nothing coming out on pin 4, which should be a simple 2:1 divider outputting 1/2 the reference clock. Unless it gets turned off because the PLL is detecting funky inputs. Which seems like it would be a really bad idea to me.

What I should probably do, though, is to rig a 5 MHz signal into pin 14 on IC803 and see if that "wakes up" the microcontroller.

 

[TM86]

Rigged a 5 MHz signal into the base of Q803, which feeds IC803 pin 14 via it's collector. Microprocessor started working immediately. Took it away, microprocessor halted. I think IC802 is good.

 

[nomadradio]

Well that's just ducky. If there's no 5 MHz coming out of IC3 (pll) pin 4, it will never transmit, either. The 5 MHz is needed to "close" the loop. Won't lock to a frequency without it.

The NDC40013 isn't the same as a PLL02A chip, is it? My Lou Franklin PLL book is at work and I'm at home.

The missing 5 MHz is a show stopper for more than just the display/keyboard processor. The PLL won't lock without it.

73

 

[TM86]

The NDC40013 isn't the same as a PLL02A chip, is it? My Lou Franklin PLL book is at work and I'm at home.

No, it's a completely different beast. It was designed to replace the MC14526/MC14568 pair of chips that made up the PLL circuit in some first generation 40 channel radios. THe PLL02A dosn't have the 1/2 reference output that I can see.

But they do face each other in the 1988 version of the book.

I've got a feeler out to a gentleman in Australia who may or may not have a loose NDC40013, or a radio in otherwise bad shape with one in it. If he doesn't have anything then I need to start thinking along the lines of what guitar_199 proposed.

I sure as heck don't see anything on fleaBay right now that would be worth taking a chance on for the prices people are asking.

 

[TM86]

Still looking for a replacement PLL, but not having much luck.

Plan B is starting to move forward, though. I've got the keypad from the radio talking to an Arduino. Got lucky as the pins happened to be the perfect spacing to fit into a breadboard. Right now it's just press button, see number, but you gotta start somewhere.

And interfacing a modern MCU with the original keypad from 45 years ago is kinda neat.

 

[357]

try that freeze spray on the chip and see if it wakes up.

NDC

NDC40013 IC

www.wagneronline.com.au

not sure if thats the right chip or not

 

[TM86]

try that freeze spray on the chip and see if it wakes up.

NDC

NDC40013 IC

www.wagneronline.com.au

not sure if thats the right chip or not

Right chip, but they don't have any. If I want to do a bulk buy they'll try to get them for me through their supplier network. Maybe I'm being picky, but I can't see putting in an order for 50 or 100 (or more) chips just to fix one radio, even it did mean I cornered the market on NDC40013's.

Thanks for taking the time to find that, though. I do appreciate that you went out of your way to offer some help.

In the meantime I've got the Arduino talking to the Si5351A module enough to generate a single frequency. Next I need to get it to change in response to keypad input. I really should have learned to code at some point, would make this so much easier.

 

[TM86]

Nobody asked, but I'm pretty happy about this.

Si5351A output now matches desired VCO frequency. I may yet double it and then feed it through a divide by 2 circuit to reduce jitter, an idea stolen from Nomad. Also thinking about feeding the output signal through the original VCO buffer and taking advantage of the minimal low pass filtering there.

Channel display works.

Scanning halfway works, need to put this in the radio and hook it to the squelch circuit for the other half.

Direct channel input works.

Using the fast and slow up and down buttons works.

And the channel 9 button works.

Still to go, Priority channel, Alternate channel, and Memory channels and the associated scan functions that go with each of those.. After that, SSB offsets and the clarifier.

Then I get to see if the rest of the radio works at all.

 

[TM86]

Another update that no one asked for.

Priority channel, alternate channel, memory channels, and scan functions are all working. Got a "prototype" squelch stop on the scanning as well. It sometimes moves forward a channel when it shouldn't. Considering everything I've had to debug that flat out didn't work on first two or three (or more) attempts I'll take it for now.

SSB offsets are in the code now. I need to see what they are on a live radio for accurate numbers. Fortunately I do have a working NDI-PC200 to check against. And no, sacrificing the NDI was never on the table.

Clarifier I need to check voltages on, I may have to convert it to use a 5V source so it doesn't blow up the Arduino. The Arduino might tolerate 8V for a while, but sooner or later exceeding specs has consequences.

Had a couple of chips in a small pile I've been dragging around since the 90's that when combined make a decent D-type flip flop, so I've got my divide by two circuit built and working as of a few hours ago. Need to do the math to not only bring up the Si5351A output up to the right numbers but also make sure the offsets aren't too far off set.

Figured out I can go cheap and use a resistor/cap combo for a low pass filter, which is probably a good thing to do before hitting the old VCO buffer transistor. I'm thinking at least three rounds of filtering that way. But I don't have the resistors or caps on hand, so those are to be ordered.

I'm thinking of feeding the VCO buffer transistor (Q2) through C11, but if anyone has a better idea I'd really like to hear it. I'm also thinking it might be a good idea to pull the collector leg on Q1, the old VCO oscillator transistor, just in case.

 

[TM86]

From the I probably shouldn't have department:

Added in some extra "features" that weren't in the original mask ROM.

Arduino sketch now has three bands of 40 channels. Lowest band is the 40 channels below 1, next is FCC channels, and then 40 channels above the FCC 40. Lowest and highest bands have no skips.

Added a +10kc function.

I also made it so going from FCC channels to the 40 above is seamless, in that the channel display reads 1 through 80 and you can directly punch in any channel. Just realized that I forgot to include a +/- 40 jump on the channel display when you manually switch bands without changing channels. Maybe tomorrow.

Yes, I know the original would show 1 through 99 on the display. Problem is 71 through 96 were below FCC channel 1 and the last three were "A" channels.

Might just add a +/- 5 kc function, if I can find a key combo that I haven't already abused too badly.

Still haven't ordered the caps and resistors. I should probably stop procrastinating and put together a list of parts I need for upcoming projects so I'm not paying 5 times in shipping what I am in parts.

Computer flashed a weird message tonight after a software update. So if I'm here tomorrow crying that I lost all my work, whatever that message was about is why.

 

[TM86]

The computer didn't die, so pretty happy about that.

I ended up doing a +5kc shift only, but it combines with the +10kc shift so I can get to the spots between the the "A" channels and the next channel up.

Parts have been ordered, arrived, and I've taken everything off the breadboard and installed it on a perfboard. Lots of point to point soldering. It's ugly, but then I'm not planning on winning any awards with this thing.

Because I'm not too bright I decided to just throw some volts at it and see what happens. It came right up and started generating a signal. Lucky for me, it was the signal I was expecting. Low pass filter could be better, but I don't want to attenuate the signal too much.

Which means now I have to either integrate this into the radio or sell the radio as parts only on fleaBay.

I guess since I've come this far, might as well keep going.