KB 381 5 valve AC/DC superhet

[ms660]

The g1 voltage in the manual was measured with a 1k OPV meter, if it was measured on a low voltage range it would have a large effect on what the actual voltage was.

-ve osc grid volts is good.

Regarding the high HT voltage, I assume you've looked at the circuit in conjunction with the 2nd paragraph in "Important Notes"

Lawrence.

[crackle]

Hi Lawrence
Yes I have seen the note about valve voltages varying depending on the actual mains voltage. That is why I had attempted to reduce the overall HT to arrive at a closer value of 140v on the OP valve anode, it was originally more like 177v.
I have remeasured the voltages again this time with an AVO 9 mk2. Here is the voltage chart, it is even more confusing because I am now getting 40v on g1 of V1 instead of the .5v listed.


It is making me doubt everything I am doing, where did I get the -12v from.
I think I am going to have to have a break from this radio chassis and go back to working on the cabinet, you never know I may get a eureka moment in my sleep.
confused
Mike

[PJL]

What are the symptoms now?

It looks OK apart from V1. g1 and g2 voltages both of which are very unlikely to be correct. All valves appear to be conducting which is a good sign although the output valve is very weak as you have already identified.

[crackle]

The only real problem is the volume, it is still very quiet, I have radios with weak output valves but they still produce enough volume to require them to be turned down substantially.
When you say the voltage readings for V1. g1 and g2 are very unlikely to be correct, do you mean errors in taking the readings or a fault in the chassis.

I feel the problem is in the OP stage, there are 2 tests I will try.
1. injecting an audio signal in via isolating capacitors.
2. test the output valve and the double diode triode on my Sussex valve tester.
At least that will give an indication as to how bad the valves are.
As the chassis for the KB 405 is out of the cabinet at the moment, I may "borrow" the speaker and transformer from that to test.

Mike

[ms660]

The manual gives a 7D3 as the output valve which is a pentode, in Post # 20 you imply that a PP3521 is fitted, that's a triode.

Lawrence.

[crackle]

Its all working now.
Silly error made by me when I re-stuffed the audio coupling capacitor from R13 C16 to the hot end of the volume track.
Found by the fact that my injected signal worked fine when connected to the volume track end but not when connected to the other end of the capacitor.

All good, plenty of volume now to fill a room, still enough with the volume control turned down to a point where the distortion is not noticeable..
Right, tomorrow is definitely a cabinet finishing day then find a place to put this one on display.

Thanks everyone

Mike

[crackle]

Good lord Lawrence, I think you are right, it came fitted in the radio. It never occurred to me to check, I saw the heater voltage and current were right and assumed as the electrode structure inside was so large it was a pentode. http://www.r-type.org/pdfs/pp3521.pdf




It is doing very well for a triode, I wonder if it is worth more than the radio, I'm going to stick it on ebay. only joking.

[PJL]

Well spotted on the wrong valve. The 7D3 looks pretty rare, the PEN36C is a possibility. Either way you should check the LT current as the 7D3 is 40V and the PP3520 is 35V. There is some info here on the Cossor version of the 7D3: http://g3ynh.info/valves/old/Data-Bo...ossor-p090.pdf

Is it the original output transformer or has that been changed to match the triode?

[crackle]

The OP transformer has a different primary resistance to what is quoted, but I will check what is fitted to my KB 405 set which is of similar design and period.
Mike

[crackle]

I have tested the OP transformers on both my KB 405 and a second KB 381 and they both have a primary resistance of around 260 ohms. So it looks like the one in this 381 was not changed for the triode.
I am now using the radio with a PEN3520, the problems I was experiencing earlier with excessive anode current with this pentode was down to HF oscillation, I have fitted a .01uf cap across the primary of the transformer, rather than fit the recommended tone corrector from the anode to ground, this has cured the excessive current and the anode current is now 15mA which is rather low but that is probably because of the slightly higher bias of -10v (-8v is recommended for the PEN3520)
As a result of this pentode having an even lower anode current than the triode, the HT on the OP valves anode is 200v, a good bit higher than that given in the service manual, but all the valves should be OK at this higher voltage, and the radio wont get that much use when there are another 100 or more KB radios to choose from.

Mike

[Boater Sam]

Job done then? Its been a pleasant read, thanks.

[crackle]

Here is a photo of an 84 year old 100k "body tip spot" resistor showing its still very accurate value today. You can just make out the spiral of what I assume is a carbon compound where the paint has chipped away.

The heater ballast resistor on this radio is mounted vertically near the back of the radio, I suppose near the back it can get maximum ventilation around it. But the heat has made the card back very brittle and hence some of the card has broken away leaving the back flopping around on one side.

This is a bit of an experiment but I have tried to repair this.
First I cut away the broken edges to give a clean right angle. The cut edges were still crumbling to dust so I used superglue (it needs to be the very thin runny type) to seal all the cut edges and the crack which ran down the back so that a strong bond can later be made with PVA glue. The super glue is absorbed into the dry card and it sets rock hard giving a firm base for the PVA.

I could not find any suitable thickness hardboard so I used a piece of thin ply to make the piece to fit in the corner. It was glued overnight and I also made a thin batten to help straighten a curve which had set in the back and strengthen the new ply I had glued in.

Next I have attempted to simulate the dimpled effect of the original surface of the back by covering it in PVA and then using a fine paint brush to stipple it to make it onto more like blobs.

I wonder what it will look like when dry and painted, hopefully a little better than the flat smooth plywood.

Mike

[Herald1360]

Looks like an early carbon film rather than carbon composition resistor. As used for the "pink band" hi-stabs which are usually still spot on too.

[crackle]

Well its not exactly a brand new back, but it now fits without flopping around and fingers are kept out.

Still got a bit of work to do on the front, top and sides of the cabinet.

Mike

[crackle]

The cabinet has now had a good polish, here it is.


The chassis is a very tight fit in the cabinet, there is only about 1/8th inch clearance between the speaker transformer mounted on the speaker and the electrolytic block on the top of the chassis.


Mike

[ms660]

Not overdone....Excellent.

Lawrence.

[HamishBoxer]

That looks superb.The phrase is like a little old lady has owned it from new.