Roberts P4D

The circuit looks fairly easy, the one in the Museum though is a bit ambiguous around the aerial tuned circuit, there are dots missing to indicate where lines join, the Trader information is better.
What's your case like.

I've got one of these. SW is pretty naff even with a huge aerial. MW and LW are good and typical of what can be expected from a battery set.

When I restored mine, SW never worked and eventually I found what was probably a production fault...a short on the SW osc coil in the coil-pack. Even after fixing that, it was never lively and I struggled to get correct coverage at each end of the SW band with the generator. Valves made no difference so I assume that it just isn't very good. Can't really see the point of SW on a portable as it needs a large external aerial to get anything....!

Restoration article here

Rich

SW is pretty naff even with a huge aerial. MW and LW are good and typical of what can be expected from a battery set.Rich

A bit like the little radio I built then, it needs a good 30 foot aerial for any SW use.

Mike

Lovely write up Rich.

I need to rob my AD3 from my 1939 Lissen and make another later.

SW on portable:
Most need an aerial. But will pickup strong stations with a curtain rail wire. Many people had no electricity even in 1949.

I have a pocket sized Yaesu VR500 and a Sony ICF2001D that are not bad on SW.

My Vidor Riviera CN379 has a loop for SW, 1949. It's not bad at appropriate band, for time of day. Though better with an aerial http://www.radiomuseum.org/r/vidor_rivi ... cn_37.html

SW is very variable.

Case
The fabric is loose on base (just needs PVA) and dirt or discolouring on rear (clean or paint).
On old woodworm flight hold on base (dark so old). Otherwise OK. Paint had flaked on perspex (still flaking) and touched up with wrong colour. Actual scale OK.

I think case will be fine.

LW & MW alignment:
I didn't read the Trader carefully enough. C23 and C22 adjust LF end of LW & MW rather than coil cores (there are none). The 6MHz end of SW is set by a 5nF fixed capacitor. The service sheet suggests checking 6MHz but offers no remedy. C5 is the cap to change.

Top three holes: LO HF end LW, MW, SW
bottom two: LO LF end LW and MW
Loop RF trimming LW & MW on frame.

Other
It's essentially a 1939 radio in design, valves, appearance and battery (AD3) (1938 in USA). The AD3 was actually obsolete the year it was released as the 1/2 volume 2/3rds weight layer cells came out that year.

I'm delighted to get it especially as I don't have the predecessors to the CN379. £25 including shipping with a dissasembled Philips L2G77B inside it for £1. The Philips just needed the broken wavechange switch fixed and card added to stop alkaline D cells shorting out.

My McMichael 463 is same valves but no SW> http://www.radiomuseum.org/r/mcmichael_463.html

More later today ...

There is maybe an earlier version of P4D?
What would this be
http://www.radiomuseum.org/r/roberts_unknown_2.html
4: DK32 DF22 DAC21 DL21
( the 2 series if Octal were originally Philips' version of 1938 Sylvania Loctal rebased.)
The P4D is
4: DK32 DF33 DAC32 DL35
(The older UK 1939 CT8 / P8A side bases and Mazda Octal are in fact identical to the 1938 & 1939 US Loctal and Octal, even same pin order, hence replacement with US octal during 1942+ production was easy)

Anyway .. IF is at 456KHz and Trader says 467KHz?

Any thoughts?

I would have used whatever frequency the Trader sheet said....

But sometimes
a) It's wrong
b) the maker changes it during production.
The IFTs are perfectly aligned and don't look fiddled with.

I'm getting inclined to leave IFTs alone unless either response is bad or evidence of twiddling.

To be honest, I don't remember adjusting the I.Fs. I definitely did a full RF alignment though.

The DAC32 is very poor and the DL35 is at about 50% or less. So I used the two Sylvania tubes from my McMichael to adjust.

The "padder" end (series caps) C22 & C23 were badly wrong, adjusting them brought in the 1800m and 500m ends of the band.

But MW RF trimmer very broad (250m / 1200kHz) & LW RF trimmer not peaking at all at 1200m 250kHz. RF loop sensitivity rubbish. Why is there PVC tape on frame coils? Remove tape. MW wire is corroded and LW looks like too much wire and is more modern with a joint.

So I replace MW wire (15 turns not too tightly spaced) and now MW peaks nicely (MW coil is used on LW). I remove 10 turns on LW, still too much. Remove 5 at a time and retest. Ooops now needs trimmer fully tight. Add 4 back. LW trimmer now peaks.

SW is lively with even 1.5m of wire, excellent with my 5.5m vertical. There are TWO L.O. trimmer peak points. One must be image and wrong? Lower capacity point the correct one according to service sheet on Radiomuseum.

Sometime I'll pick up a DAC32 and DL35 cheap. For now I might fit up a pair of Russian Rod Pentodes on octal plugs. But it's pickling up SW and RTE LW fine now on the "bad" valves.

Next the case.

Ok, now after dark (ish) and the McMichael choc a block with MW. Not a sausage on the P4D. Something isn't right.

I thought more about this...

Obviously the trimmers for RF LW and MW are for HF end of bands, 250kHz and 1200kHz. So thinking about it the LF ends, 167kHz (1800m) and 600kHz (500m) need the loop coils adjusted.

1) pushing the a coil apart in the middle will reduce its inductance.
2) The LW coil is shorted out on MW, so distance of MW loop coil from LW one varies MW inductance (closer is less).
3) On LW, the MW coil is in series, so closer increases inductance.

Also examination revealed the RF tuning cap a bit butchered at LF end. So first I straightened that out.

Next I did the MW. I put trimmer slightly more loose than 1/2 way and at 600kHz peaked the RF by moving one half of coil away from the other and back.

Then peaked the 1200kHz with trimmer.

Next I added three more turns to LW (remember it had been changed and had too many turns, hence the butchering of LF of tuner cap, I had taken too many off.) Peaked LW at 170kHz. Adjusted Trimmer at 245kHz (250kHz too close to RTE 1). Rechecked and re-adjusted coil at 170kHz. Then small adjustment at 245kHz.

Rechecked and re-adjusted coil and trimmer at 600kHz and 1200kHz on MW.
Repeat LW
Final fine adjustment on MW and LW!

RTE1 radio now very strong and R4LW weak instead of non-existent. I check on WS radio that R4 is actually on air, last few weeks during most of day Droitwich is off 🙁

I'll see tonight how it does on MW. Only some of my sets get anything on MW during daytime (Spirit is too far and weak except for really good sets with RF amp).

I've washed the case and soaked loose crumpled flap which is over 1/3 of bottom. Glue and touch up. I presume the raw hole in edge of lid isn't original?

Getting all the stronger MW (at dusk) instead of nothing. R4LW now audible. So even replacing corroded frame aerial wire with exact same number of turns the frame coils need adjusted.

So from nothing to mostly working by re-doing both frame aerials and fixing RF tuning cap. The clue was plenty of SW daytime reception (Day time if MUF high = Higher frequencies 20m to 16m = Short aerial).

I have nearly refinished case ...

That looks good Michael, what have you used to hide the scuff marks on the back.

Is yours black or 'Midnight Blue'? I thought mine was black until I started to clean it.....

My wife said
"No skill in matching that colour, it's black!"

I put spot lights on to reveal its rich Midnight blue. Most of the cloth was near bare. I "recoated" all of outside with Artist's Acrylic paint mixed to match. The inside of back is original and you can't see difference. The original photos don't show how bare it was.

After brushing on I jab a stiff flat end brush all over to have it go into fabric and this also removes brush strokes. Artist's Acyrlic (or even household Vinyl which is related) isn't that different from the 1950s plastic coating on the Leatherette cloths (Original Rexine is actually 1930s probably and different) which are variously described as Vinyl and PVC. Also still made and sold by Rachford's and others. The stuff "band" speakers are covered in isn't that different either.

I filled cutout hacked on edge of rear door with a plastic wood. I've found best way to deal with old "worm holes" (I found 3) is to cut tip of a cocktail stick, poke in as far as it goes, mark, remove cut at mark and hammer in. Filler always leaves a depression, slow and messy. I removed speaker and soaked inside with "death fluid". But the flight holes look old. I'll give a 2nd dose later.

I guess (thinking about AD3) that Roberts loved big batteries. Mullard valves / Transistors / ICs and Ever Ready Batteries, wasn't that the policy. The later transistor model taking PP8? Anyone else use a PP8? Or indeed an AD3 in a 1946 to 1951 Valve "portable".

The finished Restore

The radio works about as well on all bands as any other similar set now. Even with the less than 50% emission "original" poor DL35 it's quite usable, so I put the US type octal back in the McMichael which is all US types (though some are Brimar make).

The restored rear. I missed out fixing one bad chip on rear. I can always re do later. I found five flight holes (I missed one on filling with cocktail stick), I think all old. I did insides twice with special attention to seams/joins with "fluid".

Here is the inside, the inner of back is original, not repainted.

You can see my AD3 (which has authentically EIGHT "F" cells from 2 x 996 packs as they are only €2 each, it has 60 x AA cells, Zinc are a bit less than 1/2 original U10/"B" capacity and Alkaline AA are slightly more ), which till now lived in my 1939 Lissen All Dry Portable (same as Ever Ready). The Lissen All Dry uses valves made by Unicorns (DK1 etc 1.4V edge connect P8A /CT8 series).

Nearer coil to rear is MW and is either separated into two halves (further apart is less inductance) or closer to LW (it's shorted out on MW, so reduces inductance as MW coil is closer). The MW coil was badly corroded to black copper oxide and replaced. The LW had been replaced but too many turns (no doubt that is why LF end of RF tuning cap was so badly bent out). I fixed tuning vanes and by trial and error found number of turns for 600KHz. Conversely for LW, as MW coil is in series, moving MW coil closer INCREASES inductance and making two or more groups and separating them reduces inductance (also reduces self capacitance).

RF trimmers for 250KHz and 1200KHz and "tracking" on RF tuner vanes iteratively adjusted. Possibly the LW could be a bit better done, but it works now.

So only changes APART from coils / Tuning Capacitor fixing:
AGC 0.1uF capacitor was leaky 1970s replacement, this would just mean no AGC and full gain.
HT decoupling 0.1uF for RF was a "hidden" original paper Waxy. It would simply increase HT drain by about 1mA!
Neither of these, nor nearly worn out DL35 (the DAC32 seems not too bad now) was why the set was essentially dead on MW LW (not even RTE1 LW!). The lively Shortwave which only uses external aerial was the clue.

Obviously the case is restored, rust removed from edge of speaker frame carefully, metal mesh under speaker behind cloth de-rusted and rinsed well. A bead of glue along seam between card spacer and edge of cone outside corrugation as rust had caused it to split there in places.

So I'm quite pleased with result. But again shows that you can't assume a previous person hasn't created strange faults.

The Poor image rejection on SW is evident with Radio China and a larger aerial. It's got quieter "echo" 900kHz approx off. Hence the TWO tuning points (one "false") on the SW L.O. Trimmer.

Tips for Shortwave on Battery Sets.
1) An aerial alone will not work as there is no inherent mains RF earth (exists even if live chassis mains set). You need an Earth. But you can touch the earth terminal and that will partly work. Also a wire in the opposite direction to the aerial wire will work.
2) Even a 2.5m wire, or for strong stations only 1m wire will work, but ONLY if there is a counterpoise wire on earth or an actual Earth.
3) Best band depends on time of day.
4) Compare with a known "good" set. Some days I have thought the aerial wasn't connected, entire shortwave, especially in daytime can go "dead". Night time can be just static if daytime very short waves were good due to solar activity. Solar activity switches on bands like 18MHz to 27MHz and even VHF or UHF DX but can "kill" the lower frequency bands with noise.
5) Tune VERY slowly unless there is "bandspread". One jiggle of knob can change good reception to nothing.

Maybe some day I'll pick up an R700 and a 1930s Aladdin's lamp suitcase model and that with my R77 will do me for Roberts Models. I think the P4D is representative of all the Roberts models in that style. It's interesting to compare with the Vidor Riviera I have (similar style LW, MW, SW but with internal loop for SW and the rear cover knob is optional SW aerial socket!).
http://www.radiomuseum.org/r/vidor_rivi ... cn_37.html
There is an earlier Octal version

The earlier than CN379 with also 3 bands is CN360A or CN360 http://www.radiomuseum.org/r/vidor_3_ba ... erhet.html

But miniature valves. £16 new, the Roberts P4D was £18

The CN352 has same octal line up, but no SW.
http://www.radiomuseum.org/r/vidor_supe ... e_352.html
The CN351 also same valves but no SW
http://www.radiomuseum.org/r/vidor_rivi ... cn_35.html

A 1938 Vidor 301 / Burndept 298 would be interesting, 4 wavebands, mechanical pushbutton presets and runs on batteries.

Vidor seemed to lose the plot a bit from 1950 to 1957 I think. But still most Roberts sets command a huge amount more than comparable Vidor models today.

I'm curious as to how the Vidor personal performed compared to the 1940s US models or the Ever Ready / Marconiphone model. Or the Romac. The Romac models go for serious money, being (a) cute and (b) the first in UK and (c) Not common. Seems that it was the same Romac as was still selling Auto Accessories many years later.

Hi

I was wondering if anyone could help me please.

I have recently acquired a Roberts P4D radio, and am in the process of recapping it.

However there seems to be a resistor missing which attaches to the V4 tube.

Could someone please tell me what type of resistor I would need to use to replace the missing one (wattage and ohms resistance).

Many thanks in anticipation

Nigel

Hi Nigel and welcome to the forum. 

Reading between the lines of your post, you state you're recapping the set and discovered a resistor missing but don't know its value. This suggests to me you are likely working without service data, is this correct?

The service data shows two resistors in the feed to the V4 pentode output valve (DL35) both are connected to the little tag board to the left, as shown in Michael's photo above and the underside data plan view below. Working from the tag board bottom upwards, 1st from bottom is R8 (2Meg) and 4th from bottom being R9 (750R). Which one is missing from yours?

The data gives ohms value but not wattage, of course looking at the circuit and using ohms law one could work that out but that's equally not my strong point. I'm sure the Radio boys will be along fairly shortly with much more to add, than I. However looking at the size of the two resistors I can be reasonably sure they are no more than ½W and likely ¼W, I would err on the side of caution and fit ½W.

p.s.
If you are working without data that suggests you might be a seasoned repairer and unlikely to make mistakes, however if your are relatively new to the hobby you should always work from data.