FRG7700 display chip

[crestavega]

trying to decide whether the MSM5524RS clock/counter IC in a Yaesu FRG7700 is dead....
with reference to attached extract of circuit diag.
what is the purpose of the transistor immediately to RHS of the 3.2MHz crystal on pins 16/17 of the IC?
it seems to be a buffer in order to take-off the 3Mhz waveform to drive various other peripherls?
including
anodes of the AM/PM LED indicators ()
and
a DC-DC converter to get the 50V for the fluorescnet tube display to GLOW

does this mean that
if xtal is dead then fluorescent tube display will have zero glow?
what is Q59 doing?

also
is TP07 there specifically to scope the xtal operation?

[SiriusHardware]

I'm afraid the diagram excerpt didn't make it into your post?

I do know these display ICs are a common source of trouble and very hard to get - while yours is not proven to have failed yet, it might be worth looking around for 'FRG7700 display chip replacement' or something similar to see if someone has made a drop-in hardware replacement for the chip. It seems like the sort of thing which would be worth someone's while to do.

[Phil__G]

From (faded) memory, doesnt a flashing display indicate the the synth is out of lock?

[crestavega]

Think I forgot to click "upload"
there seems to be quite a cool youtube video demonstrating how a cheap freq counter module can be dropped in, which is really neat.

but it would be nice to get the fluorescent tube thingy working for that 1970s feel.

help gratefully appreciated.

looking at it with a clearer head this morning, I think Q59 is a buffer with a zener reference to provide a certain voltage supply on its emitter side? not sure.

T

[cmjones01]

Originally Posted by crestavega

looking at it with a clearer head this morning, I think Q59 is a buffer with a zener reference to provide a certain voltage supply on its emitter side? not sure.

Yes, Q59 is the pass transistor in a simple voltage regulator which looks like it's providing a stabilised voltage supply to the inverter which powers the fluorescent display.

Chris

[SiriusHardware]

what is the purpose of the transistor immediately to RHS of the 3.2768MHz crystal on pins 16/17 of the IC?

I think it is purely to buffer the frequency coming out of the internal oscillator on pins 16/17 of the chip, and yes, that test point is purely for measurement of that frequency. Why would you need to be able to measure and set that frequency precisely using the trimmer capacitor TC03? Because it is the clock timebase for the real time clock - if it's not set exactly right the clock will gain or lose minutes. The buffer transistor is provided so that the frequency of the oscillator can be measured without the frequency meter itself loading the oscillator and causing the frequency to 'bend' while you try to measure it.

The bottom end of R236 is NOT, as you originally thought it was, connected to the junction of R235 / C214 and the buffer transistor emitter. The line going south from R236 'jumps over' that node.

That long vertical line is just the +9V supply to the buffer transistor (via R236). It originates from the emitter of the regulator Q59, passes through a 1.2R resistor R248 and then splits in various directions, west to the oscillator formed by Q60 / Q61, south to the AM/PM LED board and east then north up to the buffer transistor.

[crestavega]

thanks so much Sirius.
yes clear(er) now.

[crestavega]

cannot see any frequency at all at TP7
have removed the crystal & checked it externally, seems to wobble at the right freq.

I see 0.5V DC across pins 16-17 of the IC
I can scope approx 1V pk-pk sine wave on pin 16, at 3.2MHz

both 16 and 17 are isolated from gnd at DC

I have checked various things (Q51 & associated resistors) that might be killing the 3.2MHz

but I think it is likely to be the IC that is not driving the xtal correctly.
I note that if I probe onto pin 17 I can "clock" the display to move 1 digit along (presumably noise is getting onto the IC from the probe and acting as edges)

any one know what DC bias I should have across the x'tal? I had a vague idea if I provide that externally then I might get the xtal going properly...?

[frsimen]

The crystal oscillator would appear to be working, at least, it is working when the 'scope probe is applied to pin 16. Check that the trimmer capacitor hasn't got a dead spot, making it open circuit, as that will stop the oscillator. The additional capacitance of the 'scope probe on pin 16 could be adding enough to get the oscillator going again. It's not likely that any sort of fault in Q51 will stop the oscillator as it is only coupled via a 1pF capacitor to the oscillator's pin.

The MSM series counter ICs don't have the best reputation for reliability. The advice to look for other options given by Sirius above is worth following if checks around the crystal oscillator circuit are fruitless.

Paula

[SiriusHardware]

If you have a signal generator try running a 3.2MHz signal into each of the IC's clock pins in turn. If that perks it up, you may be able to rescue it by building a small oscillator circuit to put the original crystal in.

Rule the crystal out by fitting any other crystal of about the same frequency (3.00-3.5Mhz) - everything will run at the wrong speed of course but if a substitute crystal of roughly the right frequency makes it run, that is proof that it is the original crystal, rather than the oscillator, which is at fault.

[John KC0G]

The MSM5524RS has been mentioned a few times in the past. It was also used in the Trio R1000. Do a search for the earlier threads.

[SiriusHardware]

This one is sounding like it might just be failure of the crystal or the chip's internal crystal oscillator because it can be 'clocked' in small steps by applying the scope probe to the clock pins of the IC - or one of them.

From what CV said earlier it seems the crystal has been tried or tested in some way external to this circuit but it is possible that it is just too 'tired' to run in the original FRG circuit, but nevertheless does spark into life in another, different oscillator circuit. That's why I would like to see what happens if another crystal of about the same frequency, give or take a MHz, is tried in the FRG.

There's also Paula's suggestion that the real culprit may be one of the passive components in the clock oscillator circuit.

[crestavega]

not looking good.
I have found a 3.1MHz crystal which I stuck in, same symptoms.
removed both the trimmer and the little cap on the other side, and stuck in a pair of 10pF ceramics, similar symptoms.
now when I probe on pin 17, I get a "full-house" display coming on but without any intelligence (put the harness back on) although it does slightly change character between FREQ and CLOCK setting (some segments go off/on)
I can see the clock waveform on pin 17 but not on pin 16.
one thing left to do, remove that transistor buffer but I think this receiver is now on its way to the scrap bin. it has some provenance; it served in the radio shack of the BAS base on Signy island from 1984 - 2005
not even sure that the RF part is actually working that well. I could receive MW broadcast last night but suspiciously quiet above 2MHz

[crestavega]

Originally Posted by SiriusHardware

If you have a signal generator try running a 3.2MHz signal into each of the IC's clock pins in turn.

I did try that too. very old ADVANCE s/g but it did deliver output at around 3mHz

[frsimen]

I don't think that 10pF capacitors would be a suitable replacement for the original 47pF and 50pF trimmer. The crystal oscillator sounds like it will work properly if you fit the correct values as replacements. Replace the 47pF and try a 33pF as a replacement for the trimmer.

It would be a shame to throw away this originally quite expensive radio for what might be a relatively simple fix. If you really don't want it, perhaps you could offer it in the for sale section!

Paula

[G6Tanuki]

Originally Posted by crestavega

I can see the clock waveform on pin 17 but not on pin 16.
one thing left to do, remove that transistor buffer but I think this receiver is now on its way to the scrap bin.

Quite often these things depend on there being a clock-signal on _both_ pins of the crystal, ideally 180-degrees out-of-phase. Injecting a waveform to one pin in the absence of anything on the other just won't work.

[SiriusHardware]

If replacing the passive components as suggested above doesn't get it going, then- regarding the failure of the signal generator experiment - it may just be that the output from the signal generator - normally intended for use with receiver inputs and I.F. stages - is just not 'loud' enough to drive what is probably the input of a digital clock oscillator.

At least try building a buffered TTL oscillator using 2-3 inverter gates and apply the output from that to each of the clock pins before abandoning the radio altogether. Run the oscillator circuit on the same supply voltage as the display IC and as before, try applying the output from the oscillator to each of the display IC's clock pins in turn.

Examples of some simple TTL crystal oscillator circuits here:-

http://www.z80.info/uexosc.htm

Edit: If Tanuki's suggestion above holds true then you could add one extra inverter to the output of the oscillator to provide an inverted 'image' of the oscillator output. Feed the uninverted oscillator output to one display chip clock pin and the inverted oscillator output to the second display chip clock pin. But I would try it the simple way first, based on the observation that touching just one pin with a scope probe causes the display circuitry to advance.

[Cruisin Marine]

A simple cheap 74HC4060 HC MOS oscillator and divider would provide an alternative Oscillator. It is an extremely simple cct. to build.

It can be wired between just pins 11 and 10. Put a 10 Meg resistor across those pins, a 33K resistor from pin 10 to one side of the Xtal. The other leg of the Xtal goes straight to pin 11 of the 4060.
You will also need a say 15 pF from Pin 11 to ground and a 33 pF trimmer from the point where the 33K meets the Xtal.
Osc. out is pin 9 on the 4060.
You will also need a 100 nF decoupling cap on the supply rail supply pin to earth = pin 16.

Building that will prove the Xtal. If you choose a 74HC4060, that will run at 5 volts and supply the correct level for the FRG unit if that is TTL levels, if not, an interface would be needed, and if an inverted signal is needed you would have to add an inverter too, which is simple enough.

You may even be able to get a normal and inverted signal from pins 11 or 10 and 9, so no inverter would be needed.

[crestavega]

Originally Posted by frsimen

If you really don't want it, perhaps you could offer it in the for sale section!

Paula

Yes - rest assured that if I do come to the end of my patience with it, as is looking likely, I will list it on here.