Philips N4306 (EL3558/3578) - UK Vintage Radio Repair and Restoration Discussion Forum

ben · 14th Aug 2015, 1:10 am

I am trying to revive this late 60s unit. After sorting the mechnics (usual goo, clutch inserts etc) I am trying to repair the amp.

Sound is present, but has terrible clipping and distortion after about position 1 on the volume control. A quick look at the output stage.... I measured no dead shorts across the two output transistors but I found R68/R69 (both 3R3) burned, I believe these are emitter resistors. Replaced them, Changed c43 (220uF) at the same time whilst I was in there. No change. Removed the transistors from heatsink in case there was some tin whisker grounding things out , and tried briefly -no change. Presumably one or maybe both the transistors are faulty, but I haven't yet taken measurements with power on as I need to knock up some kind of jig - access is pretty dire on these.

Now my question is, the schematic has these as AC128, yet this machine has OC78 types. Anyone know if this was a production change or has someone been there and replaced them at some stage? It certainly looks like the latter.

Maarten · 14th Aug 2015, 3:32 am

If yours is an early model or from another factory, it could have had those OC78 from the factory. The schematic without service bulletins and version- or countryspecific information is only really representative of one version.

Is there a label on the chassis with a few letters and numbers, such as ZB or WR? The label could also be next to the model label. Is there an execution suffix behind the N4306 model number?

On the other hand, the OC78 my have already been a few years out of production.

ben · 14th Aug 2015, 11:23 am

Thanks Maarten. It's a /15 suffix (bought in UK so I think that's correct) s/n is 021411

TonyDuell · 14th Aug 2015, 9:22 pm

For what it's worth, the service sheet for my Philips AG4000 record player says that the
input transistor can be AC125 or OC75 and that the output pair are either AC128s or OC74s. Replacements should be with the same type. I can't see any R's or C's that depend on the transistor type though.

Jimmyhaflinger · 15th Aug 2015, 10:02 pm

the N4306 is from late 1967, the OCxx transistors were last made in 1964 or so i think, so i guess they were improperly replaced by someone

BTW i'm also restoring a N4306, the amp works fine in my recorder, so if you need some voltage readings let me know!

Maarten · 16th Aug 2015, 1:47 am

I think it is likely they were replaced sometime, especially since it is a recorder from the UK and it seems those transistors were last used in equipment from the UK (so probably abundantly available as surplus material).

However one small detail that I'd still like to know... Is there a separate label with two letters indicating the factory, such as WR, AH, ZB?

Jimmyhaflinger · 16th Aug 2015, 11:10 pm

i have two N4306 recorders, both are WR, while my EL3558 is AH

Maarten · 17th Aug 2015, 3:06 pm

Those make sense (Austria and Belgium, with Belgian production maybe already being phased out around the time the N4306 came to market), but I'm wondering whether there ever was production in the UK. I hope Ben gets around to having another look at his recorder.

Jimmyhaflinger · 17th Aug 2015, 9:39 pm

maybe also the early EL3578s were still made in Belgium?

by the way, while the basic mechanism is the same for both recorders, the brakes were redesigned and are a lot better in the EL3578 and N4306 as they were made from durable fiber pads, while they're made of the usual gooey rubber in the EL3558

ben · 7th Sep 2015, 2:16 am

Update. Fixed the recorder.

Upon removing the transistors completely and taking them out the heat sink I found one to be an AC128 and the other an OC78 - bit of a bad practice to have unmatched ones in the o/p stage !

The OC78 measured o/c all ways. I suppose one of them might also have shorted to the can.

Replaced with AC188, so far so good.

Now all I need is a speed change knob....

ben · 7th Sep 2015, 2:18 am

Quote:

Originally Posted by Maarten

However one small detail that I'd still like to know... Is there a separate label with two letters indicating the factory, such as WR, AH, ZB?

No sign of any such labels unfortunately Maarten.

ben · 7th Sep 2015, 12:26 pm

Correction: Found the label!
WR 02 46 67

Presumably that's 1967 as year of manufacture?

Maarten · 9th Sep 2015, 3:18 am

Week 46 of 1967. WR for the Vienna factory, design revision 02.

I think we can safely say that the OC78 was not a factory original part in any way!

Jimmyhaflinger · 10th Sep 2015, 12:12 am

Quote:

Originally Posted by ben

Now all I need is a speed change knob....

the speed change knob from a much more common N4308 fits just fine

radiotechnician · 10th Sep 2015, 1:09 am

The Canadian distribution manual shows AC128. It was know as the
'Continental 220' here.

My reference manual is TR2007-8-66.

That model was one of the series I serviced as my firm sold them.

Jimmyhaflinger · 10th Sep 2015, 10:19 pm

this recorder, along with all other models based on this mechanism, should always have their speed selector set to the 4,75 position when not in use, otherwise the capstan belt will get stretched if left in the 9,5 position and will slip or fall off the pulley the next time the recorder is used at 4,75

14th Aug 2015, 1:10 am	#1
ben Dekatron Join Date: Aug 2006 Location: Madrid, Spain / Wirral, UK Posts: 7,498	Philips N4306 (EL3558/3578) I am trying to revive this late 60s unit. After sorting the mechnics (usual goo, clutch inserts etc) I am trying to repair the amp. Sound is present, but has terrible clipping and distortion after about position 1 on the volume control. A quick look at the output stage.... I measured no dead shorts across the two output transistors but I found R68/R69 (both 3R3) burned, I believe these are emitter resistors. Replaced them, Changed c43 (220uF) at the same time whilst I was in there. No change. Removed the transistors from heatsink in case there was some tin whisker grounding things out , and tried briefly -no change. Presumably one or maybe both the transistors are faulty, but I haven't yet taken measurements with power on as I need to knock up some kind of jig - access is pretty dire on these. Now my question is, the schematic has these as AC128, yet this machine has OC78 types. Anyone know if this was a production change or has someone been there and replaced them at some stage? It certainly looks like the latter. __________________ Regards, Ben. Last edited by Brian R Pateman; 14th Aug 2015 at 5:54 am. Reason: Hyperlink removed.

14th Aug 2015, 3:32 am	#2
Maarten Dekatron Join Date: Dec 2007 Location: Haarlem, Netherlands Posts: 4,199	Re: Philips N4306 (EL3558/3578) If yours is an early model or from another factory, it could have had those OC78 from the factory. The schematic without service bulletins and version- or countryspecific information is only really representative of one version. Is there a label on the chassis with a few letters and numbers, such as ZB or WR? The label could also be next to the model label. Is there an execution suffix behind the N4306 model number? On the other hand, the OC78 my have already been a few years out of production. Last edited by Maarten; 14th Aug 2015 at 3:47 am.

14th Aug 2015, 11:23 am	#3
ben Dekatron Join Date: Aug 2006 Location: Madrid, Spain / Wirral, UK Posts: 7,498	Re: Philips N4306 (EL3558/3578) Thanks Maarten. It's a /15 suffix (bought in UK so I think that's correct) s/n is 021411 __________________ Regards, Ben.

14th Aug 2015, 9:22 pm	#4
TonyDuell Dekatron Join Date: Jun 2015 Location: Biggin Hill, London, UK. Posts: 5,208	Re: Philips N4306 (EL3558/3578) For what it's worth, the service sheet for my Philips AG4000 record player says that the input transistor can be AC125 or OC75 and that the output pair are either AC128s or OC74s. Replacements should be with the same type. I can't see any R's or C's that depend on the transistor type though.

15th Aug 2015, 10:02 pm	#5
Jimmyhaflinger Heptode Join Date: Feb 2007 Location: La Spezia, Italy Posts: 834	Re: Philips N4306 (EL3558/3578) the N4306 is from late 1967, the OCxx transistors were last made in 1964 or so i think, so i guess they were improperly replaced by someone BTW i'm also restoring a N4306, the amp works fine in my recorder, so if you need some voltage readings let me know!

16th Aug 2015, 1:47 am	#6
Maarten Dekatron Join Date: Dec 2007 Location: Haarlem, Netherlands Posts: 4,199	Re: Philips N4306 (EL3558/3578) I think it is likely they were replaced sometime, especially since it is a recorder from the UK and it seems those transistors were last used in equipment from the UK (so probably abundantly available as surplus material). However one small detail that I'd still like to know... Is there a separate label with two letters indicating the factory, such as WR, AH, ZB?

16th Aug 2015, 11:10 pm	#7
Jimmyhaflinger Heptode Join Date: Feb 2007 Location: La Spezia, Italy Posts: 834	Re: Philips N4306 (EL3558/3578) i have two N4306 recorders, both are WR, while my EL3558 is AH

17th Aug 2015, 3:06 pm	#8
Maarten Dekatron Join Date: Dec 2007 Location: Haarlem, Netherlands Posts: 4,199	Re: Philips N4306 (EL3558/3578) Those make sense (Austria and Belgium, with Belgian production maybe already being phased out around the time the N4306 came to market), but I'm wondering whether there ever was production in the UK. I hope Ben gets around to having another look at his recorder.

17th Aug 2015, 9:39 pm	#9
Jimmyhaflinger Heptode Join Date: Feb 2007 Location: La Spezia, Italy Posts: 834	Re: Philips N4306 (EL3558/3578) maybe also the early EL3578s were still made in Belgium? by the way, while the basic mechanism is the same for both recorders, the brakes were redesigned and are a lot better in the EL3578 and N4306 as they were made from durable fiber pads, while they're made of the usual gooey rubber in the EL3558 Last edited by Jimmyhaflinger; 17th Aug 2015 at 9:45 pm.

7th Sep 2015, 2:16 am	#10
ben Dekatron Join Date: Aug 2006 Location: Madrid, Spain / Wirral, UK Posts: 7,498	Re: Philips N4306 (EL3558/3578) Update. Fixed the recorder. Upon removing the transistors completely and taking them out the heat sink I found one to be an AC128 and the other an OC78 - bit of a bad practice to have unmatched ones in the o/p stage ! The OC78 measured o/c all ways. I suppose one of them might also have shorted to the can. Replaced with AC188, so far so good. Now all I need is a speed change knob.... __________________ Regards, Ben.

7th Sep 2015, 12:26 pm	#12
ben Dekatron Join Date: Aug 2006 Location: Madrid, Spain / Wirral, UK Posts: 7,498	Re: Philips N4306 (EL3558/3578) Correction: Found the label! WR 02 46 67 Presumably that's 1967 as year of manufacture? __________________ Regards, Ben.

9th Sep 2015, 3:18 am	#13
Maarten Dekatron Join Date: Dec 2007 Location: Haarlem, Netherlands Posts: 4,199	Re: Philips N4306 (EL3558/3578) Week 46 of 1967. WR for the Vienna factory, design revision 02. I think we can safely say that the OC78 was not a factory original part in any way!

10th Sep 2015, 1:09 am	#15
radiotechnician Pentode Join Date: Jan 2013 Location: Powell River, British Columbia, Canada. Posts: 217	Re: Philips N4306 (EL3558/3578) The Canadian distribution manual shows AC128. It was know as the 'Continental 220' here. My reference manual is TR2007-8-66. That model was one of the series I serviced as my firm sold them. __________________ Steve Dow VE7ASO

10th Sep 2015, 10:19 pm	#16
Jimmyhaflinger Heptode Join Date: Feb 2007 Location: La Spezia, Italy Posts: 834	Re: Philips N4306 (EL3558/3578) this recorder, along with all other models based on this mechanism, should always have their speed selector set to the 4,75 position when not in use, otherwise the capstan belt will get stretched if left in the 9,5 position and will slip or fall off the pulley the next time the recorder is used at 4,75