Royce Sealed PLL Schematics

[brandon7861]

Here is the schematic I made for the Royce 1-621 PLL unit. I hope it helps someone.

Also attached is a PDF I found online for the 1-632 that contains a schematic for that PLL unit of that radio.

If anyone knows definitively whether or not the 641 PLL unit is the same as the 632 PLL unit, I would sure like to know. Thanks!

 

[brandon7861]

I added the parts placement for the 621 that corresponds with the part numbers I used in the schematic. The foil is reversed so the parts are positioned on page as they are looking down at the parts side of the PLL board. If you want to use that for test points, flip it so the left is on the right in a photo app that way the solder positions line up again.

 

[brandon7861]

Reposting PLL schematic. First one had a couple text errors because I wrote down 39.160MHz instead of 36.190MHz. I only noticed when going to buy a replacement 36.190MHz crystal. There was also a trace connecting the base of Q10 directly to + rail from when I was trying to reroute stuff.

 

[TM86]

If anyone knows definitively whether or not the 641 PLL unit is the same as the 632 PLL unit, I would sure like to know. Thanks!

I can't say definitively, but I did crack open a 641 tonight and had to take the shield off the PLL unit. Looks like the same parts are all in the same places. Uses a M58473P for the PLL chip.

 

[brandon7861]

This one uses the M58473P as well. I forgot to add that to the schematic.

THERES A PLL UNIT DESIGN FLAW....
I just had a breakthrough with the PLL I have been working on. First, I found that the 36.190MHz crystal was bad, wouldn't even rattle my test oscillator, so after ordering a new one from Kens (and that crystal did test good) and installing it, the 36MHz oscillator was still not working. So I took that transistor out and tested it. It tested good, but the Hfe was 49. That math just dont work with those resistor values. There was barely any voltage drop across the 1k resistor feeding the collector, so nowhere near a mA... So, I decided to temporarily hold a 220k resistor across (in parallel with) the 270k bias resistor already present, and for that brief moment, the receive came alive!!! I am not sure why they used such a large value of bias resistor having such a low Hfe transistor in there. Not suer how it even worked before.

With that mystery whooped, I couldn't be happier calling it a night. That bias resistor can be switched out tomorrow (later today). Another Royce back on the air!!!!

 

[Mud Duck 486]

Hello gentleman, I too have a 641 and a 678 that have PLL issues. The 641 will RX but not TX and the 678 does not RX or TX.. I replaced all the C460 transistors and C710s..plus tried new chips in both and to no avail.. The 641 has all the correct Freq. outputs displaying per synth alignment so wondering if that one is external? But the 678 has bad freq's for TP2 and TP3.. Haven't touched them in a while but while searching for info today came across this thread and has re-sparked my interest! Any thoughts or help would be greatly appreciated! I did pick up a new better transistor tester and would like to pull them all and retest gain for my own edification.. also never even thought about crystals could be shot.

 

[Klondike Mike]

... First one had a couple text errors because I wrote down

Well done on the schematic. Although, I think it still needs revision. I specifically eye the electrolytics. You list 2, c46 & c47. c47 is listed as 120 nf. Never seen one of those before.

The 641 PLL has 4 electrolytics. So unless you missed a couple in the 621, I don't think they are the same.

 

[brandon7861]

Well done on the schematic. Although, I think it still needs revision. I specifically eye the electrolytics. You list 2, c46 & c47. c47 is listed as 120 nf. Never seen one of those before.

The 641 PLL has 4 electrolytics. So unless you missed a couple in the 621, I don't think they are the same.

Thanks! C47 and C120 are tantalum and C51 and C46 are aluminum electrolytics. Two tantalum caps were in parallel at C120 ~ one on the back of the board making a total of three tantalums used). That makes a total of 4 "electrolytic caps" on the PLL I worked on. Did yours with 4 (assuming aluminum) electrolytics also have a couple teardrop tantalum caps or were they all aluminum electrolytics?

I no longer have the PLL and cannot verify the 120nF ratings, but I do have some pictures of the PLL. Apologies for the overlap and my inability to neatly splice them together...

 

[brandon7861]

I thought I missed a trace because I didn't see D80, but I now see I mislabeled D80 as D85 (doubling up on the use of #85, just a typo) , and thats right next to C51 which I accidentally used the wrong capacitor symbol for. I wil fix it.

 

[brandon7861]

Hello gentleman, I too have a 641 and a 678 that have PLL issues. The 641 will RX but not TX and the 678 does not RX or TX.. I replaced all the C460 transistors and C710s..plus tried new chips in both and to no avail.. The 641 has all the correct Freq. outputs displaying per synth alignment so wondering if that one is external? But the 678 has bad freq's for TP2 and TP3.. Haven't touched them in a while but while searching for info today came across this thread and has re-sparked my interest! Any thoughts or help would be greatly appreciated! I did pick up a new better transistor tester and would like to pull them all and retest gain for my own edification.. also never even thought about crystals could be shot.

The 678 sounds like the issue I had with that 621. The 36MHz oscillator went dead, and that threw the PLL out of lock and drove the VCO to its limit (around 38.6MHz if I remember correctly), and with the PLL out of lock, the 10.695 oscillator is dead, so instead of a 37MHz-10MHz=27MHz output, I ended up having 38.6MHz coming out of it. Getting the 36MHz oscillator running fixed that.

If the 641 has the correct frequencies coming out of those test points, its not the PLL unit.

Edit: I would slow down with blindly replacing transistors simply because you can. Many times, component values are specifically chosen based on the characteristics of the part (which can change from batch to batch for the same # part). For example, crystal oscillators usually run around 1mA of collector current. Take a look at my PLL schematic. The lower left section is the 36MHz oscillator that I had trouble with. When I took a voltage measurement across R62 (the 1kΩ collector resistor), I had only a small fraction of a volt there. With 1mA going through a 1kΩ resistor, there should be 1v across it. There wasn't. Thats when the transistor came out and the low gain hinted at a failing transistor or bias problem. In my case, the transistor was making no unexpected noise, so a little more bias current made it work. Now, did the previous owner put the wrong value base resistor in there or was the transistor actuially failing? It arrived broke, and screwdrivered, so I don't know.. What I do know is that the oscillator is now running at 1mA and the new (old) crystal should have a chance to last a while. My point is that if someone goes in there and swaps out that 46 Hfe part with one having a Hfe of 360, the crystal may not be too happy about that.

 

[Mud Duck 486]

The 678 sounds like the issue I had with that 621. The 36MHz oscillator went dead, and that threw the PLL out of lock and drove the VCO to its limit (around 38.6MHz if I remember correctly), and with the PLL out of lock, the 10.695 oscillator is dead, so instead of a 37MHz-10MHz=27MHz output, I ended up having 38.6MHz coming out of it. Getting the 36MHz oscillator running fixed that.

If the 641 has the correct frequencies coming out of those test points, its not the PLL unit.

Edit: I would slow down with blindly replacing transistors simply because you can. Many times, component values are specifically chosen based on the characteristics of the part (which can change from batch to batch for the same # part). For example, crystal oscillators usually run around 1mA of collector current. Take a look at my PLL schematic. The lower left section is the 36MHz oscillator that I had trouble with. When I took a voltage measurement across R62 (the 1kΩ collector resistor), I had only a small fraction of a volt there. With 1mA going through a 1kΩ resistor, there should be 1v across it. There wasn't. Thats when the transistor came out and the low gain hinted at a failing transistor or bias problem. In my case, the transistor was making no unexpected noise, so a little more bias current made it work. Now, did the previous owner put the wrong value base resistor in there or was the transistor actuially failing? It arrived broke, and screwdrivered, so I don't know.. What I do know is that the oscillator is now running at 1mA and the new (old) crystal should have a chance to last a while. My point is that if someone goes in there and swaps out that 46 Hfe part with one having a Hfe of 360, the crystal may not be too happy about that.

Thanks for the info! Yeah, I've never dived into fixing a radio with PLL problems before so still learning about gains and how they affect the crystals.. I had tested the gains of my transistors and found some where well below their HFE manufactured range which is why replacement was occuring. I figured putting better gains in would make it happy but maybe I knocked it further out of whack at the other end of gain spectrum.. So, from what I have in my notes here about the 678 is this: TP1 was good at 10.24Mhz, TP2 was at 61.7Mhz, and TP3 only had 26.965 (CH.1) on TX only. I wrote Junk for RX freq. which I believe I checked all channels and no one lined up. I think I maybe replaced a few things and wrote a date of 10/11/22 that the TP2 was at 37.66Mhz? But I know for a fact I worked with it after and that value is back at 61.7 again. So maybe some intermittent life was given to someone.. But even when it was good the radio was still dead. Your thoughts?

 

[brandon7861]

Thanks for the info! Yeah, I've never dived into fixing a radio with PLL problems before so still learning about gains and how they affect the crystals.. I had tested the gains of my transistors and found some where well below their HFE manufactured range which is why replacement was occuring. I figured putting better gains in would make it happy but maybe I knocked it further out of whack at the other end of gain spectrum.. So, from what I have in my notes here about the 678 is this: TP1 was good at 10.24Mhz, TP2 was at 61.7Mhz, and TP3 only had 26.965 (CH.1) on TX only. I wrote Junk for RX freq. which I believe I checked all channels and no one lined up. I think I maybe replaced a few things and wrote a date of 10/11/22 that the TP2 was at 37.66Mhz? But I know for a fact I worked with it after and that value is back at 61.7 again. So maybe some intermittent life was given to someone.. But even when it was good the radio was still dead. Your thoughts?

If TP1 looked good, and TP3 looked good (as it should only be active during TX and it had the right freq), something associated with only TP2 seems to be the the issue..

For there to be the proper signal at TP3 during TX, the VCO should be functioning as its signal is mixed with the 10.695 before leaving TP3, and you had that. During RX however, the 10.695 osc and mixer to TP3 is turned off and the VCO signal goes to the now active VCO amplifier that provides the external first RX mixer with its LO signal. If your notes are accurate, I would think the problem is around that VCO amplifier stage. You could start by testing the voltages associated with it during RX and see if things look good.

BUT FIRST>>>Pin 22 should have 9v on it during RX (this is the signal that tells the PLL unit whether it is supposed to be doing RX or TX stuff). Check to see if you have 9v on pin 22 (RX) first. If you do have 9v, then see what the voltage is at the emitter of (what I call) Q97, the VCO amplifier.

 

[Klondike Mike]

... The 641 PLL has 4 electrolytics. So unless you missed a couple in the 621, I don't think they are the same.

• The 641 PLL has 4 aluminum electrolytics.
• After reviewing your photos, the PLL modules are definitely not the same.
• I do not have the 641 PLL on hand and am only relying on a photo.
• The 120nf value is valid for a Tantulum. Aluminum electrolytics do not have values that low hence my questioning.

 

[brandon7861]

• The 641 PLL has 4 aluminum electrolytics.
• After reviewing your photos, the PLL modules are definitely not the same.
• I do not have the 641 PLL on hand and am only relying on a photo.
• The 120nf value is valid for a Tantulum. Aluminum electrolytics do not have values that low hence my questioning.

Yes, you are right, the 641 has the sideband PLL. It's the same as the 632 PLL (first post, 1-632 attachment, at bottom of PDF)

 

[Mud Duck 486]

If TP1 looked good, and TP3 looked good (as it should only be active during TX and it had the right freq), something associated with only TP2 seems to be the the issue..

For there to be the proper signal at TP3 during TX, the VCO should be functioning as its signal is mixed with the 10.695 before leaving TP3, and you had that. During RX however, the 10.695 osc and mixer to TP3 is turned off and the VCO signal goes to the now active VCO amplifier that provides the external first RX mixer with its LO signal. If your notes are accurate, I would think the problem is around that VCO amplifier stage. You could start by testing the voltages associated with it during RX and see if things look good.

BUT FIRST>>>Pin 22 should have 9v on it during RX (this is the signal that tells the PLL unit whether it is supposed to be doing RX or TX stuff). Check to see if you have 9v on pin 22 (RX) first. If you do have 9v, then see what the voltage is at the emitter of (what I call) Q97, the VCO amplifier.

So luckily I have the radio out right now and just tested PIN 22 for 9V. It started with 9.12volts and then slowly is draining off and just continues to.. I had noticed this previously and was wondering if the 9v regulator circuit is compromised.. Who controls the 6 & 9 volts in these radios?