Marconi CR300 again - UK Vintage Radio Repair and Restoration Discussion Forum

Malcolm G6ANZ · 20th Apr 2020, 11:53 am

I'm not sure if this should go under 'success stories' or here. Anyway, my CR300 is now functioning. I finally got the drum stringing completed and it all turns as it should although I made one error which was countered by another. I got a single twist in the string running from the drum to the switch and I also managed to wind the string around the switch drum the wrong way. I only found this out when I turned the receiver the right way up and could see the twist. In hindsight the string went round the switch drum the wrong was as I was rotating the switch each time I made a turn to make sure the drum was turning the right way, which it was. As it was such a swine to do it's staying.
I had already recapped the PSU and proved that it was working correctly. The wiring under the beast is a rats nest so I took resistance measurements from the output valve anode to the output transformer and all the transformer primaries and secondaries. These were ok. I then put a signal from an audio generator across the primary and there was a sound from the speaker. So the output stage should work. There was a resistor from the transformer to the HF gain pot that was out of spec as well as having a dry joint at one end this was replaced. The grid coupling capacitor looked in good condition and didnt test leaky, so I powered it up. To be greeted by silence. Only a crackle from the speaker as the wave change was operated and a bit of noise from the volume pot. By measuring the values of the HF gain, volume and desense pots I found that they were all noisy or of the wrong value. The volume pot was opened up and cleaned. The desense pot had a large chunk missing from its track and the HF gain pot was still noisy after cleaning. I powered up again and there was some signs of life as there was hiss from the speaker. Otherwise nothing even with a signal directly on the aerial input. At this point I noticed that some of the valves were not glowing. Definite hint. The valves run run in two chains of four in parallel to run off of the 24V supply. As the output valve was glowing it seemed a reasonable assumption that the rest of the valves in that chain were OK and the faulty valve in the other chain. The valves in this chain were those for the RF/IF. Heater continuity was checked on each valve and all were OK. Switching on after replacing all the valves resulted in them all lighting up. A few minutes later there was noise from the speaker and I could tune in the sig gen. After a bit of playing around There came through a Russian station.

Malcolm

g4aaw pete · 20th Apr 2020, 12:08 pm

Well done Malcolm, tenacity wins the day.
These are something of an oddity, and certainly an interesting project.

I’ve always wanted one of these, already having had plenty of CR100s & CR150s. However, a house move is planned when the current situation changes, so it’s fortunate I haven’t got one.

Malcolm G6ANZ · 20th Apr 2020, 2:46 pm

Thanks Pete. I never expected it to work as it had been in a garage for a good number of years as the crud on the chassis shows. All the wiring is lacquered cloth so hasn't rotted out. The only bits that look a bit dodgy are the terminations of the screened wires.
Next job is to check it for out of spec R's and C's. I expect there to be a number of these. I will change any out of spec items one at a time and make sure the set carries on working after the change.

Malcolm G6ANZ · 23rd Apr 2020, 6:45 pm

I've started looking into which R and C are out of spec. I started with the 6V6 output valve, mainly as it's easy to access. Someone in the distant past has removed the screen grid resistor and replaced it with a piece of wire. Definitely not original as it is PVC covered and only tacked on the terminals.
I will replace it with the correct resistor but is there a reason for cutting it out?
The audio side works ok so the valve is all right.
Malcolm

g4aaw pete · 23rd Apr 2020, 8:38 pm

I note that the CR100 has the KT63 / 6V6 screen going straight to the HT side of the o/p transformer. Perhaps the previous owner though there'd be some advantage in not having a screen resistor?

Malcolm G6ANZ · 23rd Apr 2020, 9:29 pm

What would there be the advantage in not having a resistor? Apart from keeping the screen grid at a lower voltage than the anode.

g4aaw pete · 23rd Apr 2020, 10:20 pm

Quote:

Originally Posted by g4aaw pete

I note that the CR100 has the KT63 / 6V6 screen going straight to the HT side of the o/p transformer. Perhaps the previous owner thought there'd be some advantage in not having a screen resistor?

I'd say it was a mis-guided thought!
Can't see any advantage. Perhaps a wiring error during a repair attempt.

turretslug · 25th Apr 2020, 10:48 am

Out of curiosity (and quiet times....) I revisited the circuit. The output valve's screen grid feed is listed as 4k7, with no associated bypass capacitor. That's quite high for what would appear to be acting as a screen-grid stopper resistor but who knows what might have had to be done in development.

I was also struck by the 1st AF amp's load resistor, a series combination of 10k and 47k. (10k to HT). This component is hardly a critical value situation, and surely a 56k could have been used anyway if the situation demanded it. In perhaps 80%+ of examples of this particular circuit ever made and used in a zillion basic radio AF amps, the resistor junction is decoupled to circuit 0V via a capacitor whose value varies widely but 2-8uF is a ballpark, as this stage is particularly vulnerable to HT ripple. In the case of the AF output stage and its 4k7 screen feed, a similar value capacitor from screen grid to 0V would also assist in guarding against ripple- whilst the high anode impedance of a tetrode or pentode helps with hum rejection, the screen grid can act as quite an effective modulating electrode so can demand further filtering, hence the "one stage down" sometimes seen in screen-grid vs. anode HT supply filtering arrangements. Perhaps the designer had originally added decoupling capacitors in both positions discussed, but it was found that the vibrator supply was better than expected as regards ripple and the capacitors were simply left out in production to save a few bob. A comms receiver's output transformer probably has limited response at 100Hz when used on mains as opposed to DC battery via higher frequency vibrator anyway.

Another circuit anomaly spotted (there may be others!) is that the second IF amp's cathode is switched open-circuit to DC in the "narrow" selectivity position- surely a drafting error?

Malcolm G6ANZ · 30th Apr 2020, 6:01 pm

I've been making further investigations to this radio while waiting for a selection of 1W resistors to arrive First, although it worked, the measured HT was way low. It is 120v as opposed to the 350 v it should be. The voltage from the PSU when not connected to the radio is 350v so something in the set is dragging the voltage down.
Secondly I checked the top cap voltages (grid) on the valves. mainly because they are easily accessible. All were at 0.0v or slightly negative except for the top cap on V5 which is at +1.6v. V5 (DH63) is the double diode triode detector/ amplifier.
I'm thinking of C91 being leaky. This is,as far as I can see, a flat wax covered cap.
Obviously there are several leaky capacitors dragging down the HT, finding them is not easy as the under chassis wiring is a real rats nest.

ms660 · 30th Apr 2020, 6:20 pm

That grid voltage on V5 seems low if you are using a high impedance meter.

Lawrence.

Malcolm G6ANZ · 30th Apr 2020, 6:40 pm

I am using a high impedance meter. I have always thought that a grid should not be positive WRT the chassis.

ms660 · 30th Apr 2020, 6:50 pm

Quote:

Originally Posted by Malcolm G6ANZ

I am using a high impedance meter. I have always thought that a grid should not be positive WRT the chassis.

The grid shouldn't be +ve WRT cathode, it's the voltage developed across R42 (680 Ohm) that determines the bias voltage. The grid is returned to the junction of R42 and R44 (2.2k) via the volume control and according to the manual the cathode current is approx. 1.5mA which would mean the voltage across R44 would be approx. 3.3 Volts, and the bias voltage developed across R42 would be approx. 1.02 Volts.

EDIT: Having said that I've just noticed that you said the HT is low anyway, so the voltage on the grid WRT chassis will be low.....Sorry I missed that bit.

Lawrence.

Malcolm G6ANZ · 30th Apr 2020, 7:00 pm

Thanks Lawrence. That makes perfect sense. I'll try to measure the voltage across R42 and R44. Though as you can see from the photos they could well be buried deep

Malcolm G6ANZ · 13th May 2020, 6:31 pm

I'm getting results from the receiver now. Having changed several out of spec resistors it's a lot more lively. I still have a big problem. The HT voltage from the PSU when off load is 370v but when loaded with the receiver this drops to 145v. My first thought was one of the 0.1uF HT decoupling caps was leaking, but none of then were getting warm. So I then measured the HT current drawn by each stage. Each HT feed is via a 470 ohm resistor and the manual gives a current for each one. So with Ohms law I could work out the volt drop across each resistor. If this voltage was too high then that would be the faulty section. The volt drop across each resistor was less than calculated, I put this down to the low HT volts, but it did show that there was not much wrong with the majority decoupling caps.
I next measured the total HT current by breaking into the HT feed immediately after the power switch on the main receiver. This switch switches off the HT and LT to the set. The current measured was 24mA. This appears reasonable. I then went back the PSU. Again the HT measured here on the cathode of the rectifier (6x5g) fell from 370v to 145v. In case there was a problem in the smoothing after the rectifier I measured the current going to the reservoir capacitor and the rest of the filter circuit. This was 24 mA. The AC voltage on the secondary of the transformer measured at approx 350 v and stayed constant weather the receiver was on or off.
24 mA is well within the current capacity for the rectifier so would I be correct in thinking that the rectifier is loosing emission and now, after 60 years has finally decided its had enough?

Malcolm

Silicon · 13th May 2020, 6:51 pm

The easiest way to find out is to temporarily replace the valve rectifier with a pair of silicon diodes and a surge limiting resistor.

Another possibility is a faulty connector in the rectifier valve holder which may turn it into a half wave rectifier. A faulty reservoir capacitor might give these symptoms.

ms660 · 13th May 2020, 7:54 pm

Once I had replaced the usual in the power supply the power supply I had was fine, next thing to go was the 1uF in the receiver, it turned into a low resistance VDR, ended up replacing all the usual suspects.

If you should decide to give the Osc/RF a tweak remember that the oscillator frequency is below the signal frequency on the high band.

Mine worked ok once I sorted everything out, long gone now though.

Lawrence.

20th Apr 2020, 11:53 am	#1
Malcolm G6ANZ Octode Join Date: Feb 2006 Location: Bristol, UK. Posts: 1,041	Marconi CR300 again I'm not sure if this should go under 'success stories' or here. Anyway, my CR300 is now functioning. I finally got the drum stringing completed and it all turns as it should although I made one error which was countered by another. I got a single twist in the string running from the drum to the switch and I also managed to wind the string around the switch drum the wrong way. I only found this out when I turned the receiver the right way up and could see the twist. In hindsight the string went round the switch drum the wrong was as I was rotating the switch each time I made a turn to make sure the drum was turning the right way, which it was. As it was such a swine to do it's staying. I had already recapped the PSU and proved that it was working correctly. The wiring under the beast is a rats nest so I took resistance measurements from the output valve anode to the output transformer and all the transformer primaries and secondaries. These were ok. I then put a signal from an audio generator across the primary and there was a sound from the speaker. So the output stage should work. There was a resistor from the transformer to the HF gain pot that was out of spec as well as having a dry joint at one end this was replaced. The grid coupling capacitor looked in good condition and didnt test leaky, so I powered it up. To be greeted by silence. Only a crackle from the speaker as the wave change was operated and a bit of noise from the volume pot. By measuring the values of the HF gain, volume and desense pots I found that they were all noisy or of the wrong value. The volume pot was opened up and cleaned. The desense pot had a large chunk missing from its track and the HF gain pot was still noisy after cleaning. I powered up again and there was some signs of life as there was hiss from the speaker. Otherwise nothing even with a signal directly on the aerial input. At this point I noticed that some of the valves were not glowing. Definite hint. The valves run run in two chains of four in parallel to run off of the 24V supply. As the output valve was glowing it seemed a reasonable assumption that the rest of the valves in that chain were OK and the faulty valve in the other chain. The valves in this chain were those for the RF/IF. Heater continuity was checked on each valve and all were OK. Switching on after replacing all the valves resulted in them all lighting up. A few minutes later there was noise from the speaker and I could tune in the sig gen. After a bit of playing around There came through a Russian station. Malcolm

20th Apr 2020, 12:08 pm	#2
g4aaw pete Heptode Join Date: Jan 2008 Location: Birchington Kent, UK. Posts: 595	Re: Marconi CR300 again Well done Malcolm, tenacity wins the day. These are something of an oddity, and certainly an interesting project. I’ve always wanted one of these, already having had plenty of CR100s & CR150s. However, a house move is planned when the current situation changes, so it’s fortunate I haven’t got one. __________________ Regards Pete

20th Apr 2020, 2:46 pm	#3
Malcolm G6ANZ Octode Join Date: Feb 2006 Location: Bristol, UK. Posts: 1,041	Re: Marconi CR300 again Thanks Pete. I never expected it to work as it had been in a garage for a good number of years as the crud on the chassis shows. All the wiring is lacquered cloth so hasn't rotted out. The only bits that look a bit dodgy are the terminations of the screened wires. Next job is to check it for out of spec R's and C's. I expect there to be a number of these. I will change any out of spec items one at a time and make sure the set carries on working after the change. Attached Thumbnails

23rd Apr 2020, 6:45 pm	#4
Malcolm G6ANZ Octode Join Date: Feb 2006 Location: Bristol, UK. Posts: 1,041	Re: Marconi CR300 again I've started looking into which R and C are out of spec. I started with the 6V6 output valve, mainly as it's easy to access. Someone in the distant past has removed the screen grid resistor and replaced it with a piece of wire. Definitely not original as it is PVC covered and only tacked on the terminals. I will replace it with the correct resistor but is there a reason for cutting it out? The audio side works ok so the valve is all right. Malcolm

23rd Apr 2020, 8:38 pm	#5
g4aaw pete Heptode Join Date: Jan 2008 Location: Birchington Kent, UK. Posts: 595	Re: Marconi CR300 again I note that the CR100 has the KT63 / 6V6 screen going straight to the HT side of the o/p transformer. Perhaps the previous owner though there'd be some advantage in not having a screen resistor? __________________ Regards Pete

23rd Apr 2020, 9:29 pm	#6
Malcolm G6ANZ Octode Join Date: Feb 2006 Location: Bristol, UK. Posts: 1,041	Re: Marconi CR300 again What would there be the advantage in not having a resistor? Apart from keeping the screen grid at a lower voltage than the anode.

25th Apr 2020, 10:48 am	#8
turretslug Dekatron Join Date: Nov 2011 Location: Surrey, UK. Posts: 4,394	Re: Marconi CR300 again Out of curiosity (and quiet times....) I revisited the circuit. The output valve's screen grid feed is listed as 4k7, with no associated bypass capacitor. That's quite high for what would appear to be acting as a screen-grid stopper resistor but who knows what might have had to be done in development. I was also struck by the 1st AF amp's load resistor, a series combination of 10k and 47k. (10k to HT). This component is hardly a critical value situation, and surely a 56k could have been used anyway if the situation demanded it. In perhaps 80%+ of examples of this particular circuit ever made and used in a zillion basic radio AF amps, the resistor junction is decoupled to circuit 0V via a capacitor whose value varies widely but 2-8uF is a ballpark, as this stage is particularly vulnerable to HT ripple. In the case of the AF output stage and its 4k7 screen feed, a similar value capacitor from screen grid to 0V would also assist in guarding against ripple- whilst the high anode impedance of a tetrode or pentode helps with hum rejection, the screen grid can act as quite an effective modulating electrode so can demand further filtering, hence the "one stage down" sometimes seen in screen-grid vs. anode HT supply filtering arrangements. Perhaps the designer had originally added decoupling capacitors in both positions discussed, but it was found that the vibrator supply was better than expected as regards ripple and the capacitors were simply left out in production to save a few bob. A comms receiver's output transformer probably has limited response at 100Hz when used on mains as opposed to DC battery via higher frequency vibrator anyway. Another circuit anomaly spotted (there may be others!) is that the second IF amp's cathode is switched open-circuit to DC in the "narrow" selectivity position- surely a drafting error? Last edited by turretslug; 25th Apr 2020 at 10:53 am.

30th Apr 2020, 6:01 pm	#9
Malcolm G6ANZ Octode Join Date: Feb 2006 Location: Bristol, UK. Posts: 1,041	Re: Marconi CR300 again I've been making further investigations to this radio while waiting for a selection of 1W resistors to arrive First, although it worked, the measured HT was way low. It is 120v as opposed to the 350 v it should be. The voltage from the PSU when not connected to the radio is 350v so something in the set is dragging the voltage down. Secondly I checked the top cap voltages (grid) on the valves. mainly because they are easily accessible. All were at 0.0v or slightly negative except for the top cap on V5 which is at +1.6v. V5 (DH63) is the double diode triode detector/ amplifier. I'm thinking of C91 being leaky. This is,as far as I can see, a flat wax covered cap. Obviously there are several leaky capacitors dragging down the HT, finding them is not easy as the under chassis wiring is a real rats nest. Attached Thumbnails

30th Apr 2020, 6:20 pm	#10
ms660 Dekatron Join Date: Apr 2011 Location: Cornwall, UK. Posts: 13,454	Re: Marconi CR300 again That grid voltage on V5 seems low if you are using a high impedance meter. Lawrence.

30th Apr 2020, 6:40 pm	#11
Malcolm G6ANZ Octode Join Date: Feb 2006 Location: Bristol, UK. Posts: 1,041	Re: Marconi CR300 again I am using a high impedance meter. I have always thought that a grid should not be positive WRT the chassis.

30th Apr 2020, 7:00 pm	#13
Malcolm G6ANZ Octode Join Date: Feb 2006 Location: Bristol, UK. Posts: 1,041	Re: Marconi CR300 again Thanks Lawrence. That makes perfect sense. I'll try to measure the voltage across R42 and R44. Though as you can see from the photos they could well be buried deep

13th May 2020, 6:31 pm	#14
Malcolm G6ANZ Octode Join Date: Feb 2006 Location: Bristol, UK. Posts: 1,041	Re: Marconi CR300 again I'm getting results from the receiver now. Having changed several out of spec resistors it's a lot more lively. I still have a big problem. The HT voltage from the PSU when off load is 370v but when loaded with the receiver this drops to 145v. My first thought was one of the 0.1uF HT decoupling caps was leaking, but none of then were getting warm. So I then measured the HT current drawn by each stage. Each HT feed is via a 470 ohm resistor and the manual gives a current for each one. So with Ohms law I could work out the volt drop across each resistor. If this voltage was too high then that would be the faulty section. The volt drop across each resistor was less than calculated, I put this down to the low HT volts, but it did show that there was not much wrong with the majority decoupling caps. I next measured the total HT current by breaking into the HT feed immediately after the power switch on the main receiver. This switch switches off the HT and LT to the set. The current measured was 24mA. This appears reasonable. I then went back the PSU. Again the HT measured here on the cathode of the rectifier (6x5g) fell from 370v to 145v. In case there was a problem in the smoothing after the rectifier I measured the current going to the reservoir capacitor and the rest of the filter circuit. This was 24 mA. The AC voltage on the secondary of the transformer measured at approx 350 v and stayed constant weather the receiver was on or off. 24 mA is well within the current capacity for the rectifier so would I be correct in thinking that the rectifier is loosing emission and now, after 60 years has finally decided its had enough? Malcolm

13th May 2020, 6:51 pm	#15
Silicon Nonode Join Date: Jun 2007 Location: Coulsdon, London, UK. Posts: 2,162	Re: Marconi CR300 again The easiest way to find out is to temporarily replace the valve rectifier with a pair of silicon diodes and a surge limiting resistor. Another possibility is a faulty connector in the rectifier valve holder which may turn it into a half wave rectifier. A faulty reservoir capacitor might give these symptoms.

13th May 2020, 7:54 pm	#16
ms660 Dekatron Join Date: Apr 2011 Location: Cornwall, UK. Posts: 13,454	Re: Marconi CR300 again Once I had replaced the usual in the power supply the power supply I had was fine, next thing to go was the 1uF in the receiver, it turned into a low resistance VDR, ended up replacing all the usual suspects. If you should decide to give the Osc/RF a tweak remember that the oscillator frequency is below the signal frequency on the high band. Mine worked ok once I sorted everything out, long gone now though. Lawrence.