Roberts RT7 - problem

[stonehopper]

I think you meant 'R' 27 not 'C', but I've just tried the R26 & R27 on the 200 range in the resistance test, and got no reading on the R26, and 72.9 on R27.

Checked the capacitor on the 200uA, and still zero.

Derek

[Station X]

He did mean R27 and I've corrected that.

Your results for R26 and R27 are what I'd expect. They confirm that R27 is closer to 80 ohms than 51 ohms. Replace it with 47 ohms, 56 ohms or two 100 ohm resistors in parallel. It's your choice, but I'd go for 47 ohms myself.

C32 seems to be OK, so I can't explain why it gets hot. Perhaps it's absorbing heat from another source?

[smiler411]

As far as I can see the base bias voltage on TR6 and TR7 is governed solely by R26/27 and any base current drawn by TR6/7.

I agree, only R26 and R27 affect the output stage bias.

If R27 has risen to 80 ohm from 51 then this is the cause of your problem.

correct bias voltage ignoring nbase current would be 51/2200 * 9 = 0.208V assuming a 9V battery

with values you have 80/2530 *9 = 0.2845V

You actually measured this voltage earlier and it was 0.28V with 9.7V battery.


This increase in voltage will increase current through transistors. This will mean they will start to get hot and then you enter a thermal runaway situation where less bias is needed etc so everything rapidly overheats.

provided transitors are not damaged changing R27 to 56ohm will give

56/2530 *9 = 0.199V which is very close to correct value. So fit 56ohm resistor in place of R27 and everything should work provided the transistors are not damaged.

Jeff

[stonehopper]

Originally Posted by Herald1360

Me again....
(snip)
Doesn't it have a transistor check function? The small round black socket on the left, used with the switch set to hFE.
(snip)
Chris

Yes it does, and I've just tried my OC81's in it. One came up with a reading of 80, but if I touched the top of the can with my finger it drew down to below 60.

The other reading was 130, with no effect of the finger touch.

Derek

[Herald1360]

Good, sounds like the transistors are probably OK. I tried a few (including an OC81, much to my surprise) and found that the meter seems to work, but the socket is a bit iffy and readings could vary quite a bit with finger pressure etc. I think just doing the diode checks c-b, e-b and c-e is probably easier to prove you still have a transistor rather than a blob of fused semiconductor in the can.

I suspect smiler may be right about the 80 ohms causing the problem, especially if the OC81s are reasonably high gain, though the calculations aren't quite right- the open circuit base voltage is Batt volts * 80/(80 + 2530) so it will be a little lower than stated at 0.276V for 9V batt volts.

If you reassemble everything, and connect to the battery via the meter on 200mA range, short out R27 and switch on, current should be low- a few mA. briefly take the short off R27, if the current now rises much above 10mA, R27 IS the culprit. In which case if you've got a suitable 47 to 56 ohm resistor, replace R27. (Or connect about 120 ohm in parallel).

Chris

[stonehopper]

Thanks Chris, I'd agree the diode checks seem more reliable than plugging it into an iffy socket and having one number displayed.

Question: - never having bought resistors before, am I looking for 2watt 5% carbon film 47 ohm and 2.2k ohm resistors such as those in Cricklewoods listing as 2W47R & 2W2K2?

There may be other issues to tackle yet - transformer possibly - but this is a start.

Cheers - Derek

[Station X]

2 watt resistors are really a bit over specified for the job. However as long as there's physically enough space to fit them they'll be fine.

[stonehopper]

What are my options then - the ones in there are about 12mm long? Would lower wattage be more suitable?

Derek

[Station X]

PM me your address and I'll send you some suitable resistors free of charge.

[stonehopper]

Originally Posted by Station X

PM me your address and I'll send you some suitable resistors free of charge.

A Star! I'll PM.

I'm out today, so not much else from me just now (I hear sighs of relief!)

Derek

[geofy]

Originally Posted by Station X

Are we singing from the same hymn sheet? R27 isn't the volume control. R23 is the volume control.

As far as I can see the base bias voltage on TR6 and TR7 is governed solely by R26/27 and any base current drawn by TR6/7.

Yes didn't have the circuit in front of me, not a very good idea to have the volume control as part of the bias, if the track gets worn it could mess the bias up. Most designs would have it at the input to TR4, and even here the tone control is part of the base bias for TR4.

Geof

[Station X]

Resistors in the post.

[stonehopper]

I'm back.
Further to the collector emitter connections; I have traced the OC81's (TR6 & TR7) on the board and can confirm that the collectors both go to one side of the transformer T2 but to different terminals, and also that the TR7 collector goes to the base of wax C33. Whereas the TR6 collector also goes through R30 and on to the top of wax C33.
The TR6 & 7 emitters both go to different terminals on the other side of T2.
The TR6 & 7 base are connected to different terminals on one side of the T1 driver transformer.

These shots of the base make it look pretty grim. But my camera seems to pick up things that I cannot see with the eye - even with two pairs of glasses on!
Some of the holes are unused, checking with the service sheet shows no components in those locations. What looks like solder spatter possibly is, but the whole board seems to be lacquered over - spatter and all. The repaired track is far left - first inside the boundary track.
The empty holes far left, top left, are the removed TR6,7, & C32, there are two more where the R26 & R27 have been lifted one end.

PS The orange wire floating around was removed for access to R26.

[AlanBeckett]

Ah, it takes you back. I bet it was dip soldered!
There appear to be a couple of dry joints on the first picture.
Alan

[stonehopper]

Question: These OC81m's that took an exception to 'procedures' and promptly failed, with their little rubbery jackets. It is being suggested that AC128's (half the price) are a more reliable replacement. Given that the originals were held in a copper clip riveted to the frame and therefore a decent heat sink, would the forum suggest that these AC128's might indeed be a better choice?

When the replacement OC81m's arrived, I had my doubts about them being suitable solely on the grounds that the loss of a heat sink was a course set for failure. Opinions most welcome.

Derek

[Steve_P]

Well the OC81 and OC81D both have a TO1 base.

This is a good reference guide:

http://www.vintageradio.me.uk/info/germanium.htm

An AC128 will go in it's place, but I'd rather not complicate matters.

First. change those resistors and also, if you haven't done already, C32 and C25. Without the output transistors in, Measure TR5:

Base is 1.69 Volts. Emitter is 1.6 Volts. Csollector is 8.7v. Give or take.

TR6 and TR7 Base connections should be 0.2v. Collectors should be 9v. Emitters go to earth via Transformer T2, so should read 0 Volts.

If not, Do NOT put any transistors in. Tell us what you do find.

I don't know whether your transistors are OK. might be worth getting new ones. Put these in the right way round and then connect your meter as an Ammeter in series with the battery and the radio.

Switch on and watch the current carefully. If it shoots up or something gets hot, off quick. If not, take meter out and try radio.

Cheers and Good Luck,

Steve P.

[stonehopper]

Hi again:-

Originally Posted by Steve_P

(snip)
First. change those resistors and also, if you haven't done already, C32 and C25. Without the output transistors in, Measure TR5:

Base is 1.69 Volts. Emitter is 1.6 Volts. Collector is 8.7v. Give or take.

TR6 and TR7 Base connections should be 0.2v. Collectors should be 9v. Emitters go to earth via Transformer T2, so should read 0 Volts.

Just some clarification Steve - I will need to put the TR6 & TR7 back in if I am to test for voltage - yes?

But then you say:

If not, Do NOT put any transistors in. Tell us what you do find.
Steve P.

How should I test for voltage of the TR6 & TR7 out of the set?

The new resistors R27 & R27 are fitted, with the gold band in the same position as the old (does this matter?), but have stuck with the C32 & C25, as they show no leakage on a continuity test. (Though I wish I had as the connections are frail!)

I have been down to Cricklewood and obtained a pair of AC128's. They are smaller than the original OC81's, and unlike the OC81M's have not a rubber jacket, but are metal TO1 bodies with a metal tang in place of any paint mark.

I have measure the AC128's with a 20k Ohm range and compared the readings with the OC81M's (seemingly failed), and find them thus:
AC128/ OC81
+E to -C = open circuit / 2.3
-E to +C = 2.32 / open circuit
-E to +B = open circuit / open circuit
+E to -B = 3.18 / 3.67
+C to -B = 3.18 / 3.31
-C to +B = open circuit / open circuit

I am puzzled (not too difficult) to know why the first set of readings between the collector and emitter are opposite, when all the other checks are similar.

It was suggested I had put the replacement OC81's in the wrong way round, but had been assured that the grey line on the jacket lined up with the collector.

I am now taking it that the metal tag on the AC128 is aligned with the collector (no paint). Edit: Interestingly, the triangular pattern the wires make from the bottom of the AC128's lines up with the pattern of holes in the board, whereas those on the replacement OC81M's were opposite.

Before I commence with refitting the AC128's as TR6 & TR7, I would like some guidance as to the above guys - thanks in anticipation,

Derek

[Station X]

Derek.

Resistors are not polarity sensitive. The position of the gold band is immaterial. It justs shows the tolerance of the resistor. The other bands indicate the resistor value. Red, Red, Red is 2.2K. Yellow, Violet, Black is 47R or 47 ohms. This replaces the original 51 ohm resistor.

MEASURING THE BIAS ON THE OUTPUT STAGE.

All components fitted except TR6 and TR7.

Meter set to a suitable DC volts range. If unsure start with the highest range and work down.

Red lead to chassis. Black lead to junction of R26 and R27.

Switch on and note reading. Should be about 0.2V or 200 millivolts (mV).


CHECKING VOLTAGES AROUND TR5

All settings as above, but connect black meter lead to each lead of TR5 in turn.

Check readings against service sheet.


MEASURING CURRENT DRAWN BY SET.

Meter set to a suitable milliamp (mA) range. Note that some meters use different sockets for current readings.

Disconnect positive battery lead from battery.

Connect red lead of meter to positive of battery.

Connect black lead of meter to positive battery lead going to set.

Switch on and note reading.



Report results here and we'll tell you what to do next.

[Tim]

I ordered some AC187/88s from CPC a few years ago. These had a tag which turned out to indicate the EMITTER(rather than the piant dot we are used to seeing by the collector). However the triangular pinout was as expected.

[Station X]

I've just checked "Towers" In all cases where a transistor has three leads and a tab on the can the tab is always the emitter.