Ferranti A1016 bluetooth project - Page 2 - UK Vintage Radio Repair and Restoration Discussion Forum

Boater Sam · 3rd Jul 2018, 4:29 pm

1:1 RS transformer.
I doubt the heater winding will have any problem supplying 50mA extra, all the valve filaments together could be that far out anyway.

Ambientnoise · 3rd Jul 2018, 4:41 pm

If you use a half wave rectifier, you will retain a common ground and not need a coupling transformer?

Ken

egerton · 3rd Jul 2018, 10:51 pm

I don't think I could get enough DC out of a 1/2 wave on the heater supply. I reckon on 6.3 x 0.45 = 2.8V (less allowing for diode fwd drop?). My full wave gives (6.3 x 1.414) - (2 x 0.7) = 7.5V and I just about get enough headroom to drive a regulator with some decent smoothing capacitor.

egerton · 3rd Jul 2018, 11:03 pm

Quote:

Originally Posted by Boater Sam

1:1 RS transformer.
I doubt the heater winding will have any problem supplying 50mA extra, all the valve filaments together could be that far out anyway.

Thanks, I have ordered similar. I note most are 600 ohm. Not sure if that matters for my application, but I'll see how I get on with it.

McMurdo · 3rd Jul 2018, 11:16 pm

one of these will contain 2 line-level audio transformers.

http://cpc.farnell.com/unbranded/b07...op%20isolator#

The beauty is that the plastic casing can either be retained or split open as it's only held together by its sticky labels.

Ambientnoise · 4th Jul 2018, 11:09 am

The half wave rectifier will actually give a higher peak rectified voltage (one less diode) but of course will need about 4 times the smoothing cap because of the higher ripple at 50Hz versus 100Hz. You might also consider a low dropout regulator, several good chips or diy !

Ken

egerton · 4th Jul 2018, 2:16 pm

hmmm not convinced. Anyway, I don't want to have a large smoother as it will increase switch on surge current. I already have 500 uF on the full wave to keep the rail from dipping and I'd need even more on half wave. Thanks - Paul

4th Jul 2018, 2:30 pm

Yes, half wave will be fine, say 6.3V RMS, that's nigh on 9V peak, less diode drop and using 1000uF of smoothing and you will have at least 8V to feed a 7805 regulator. Do put 10nF or so across the diode to stop switching noise.

A quick simulation at 50mA gives a minimum voltage of just over 7V, more than enough for a low drop out 7805 type of regulator. And a whole 1/2W extra from the transformer, no problem there either.

Ambientnoise · 4th Jul 2018, 3:22 pm

Switch-on surge will be a second or so and unlikely to cause problems. It will prob be limited by the transformer winding rather than the esr of the cap.
Of course, go with whatever you are comfortable with, it’s your radio !
Ken

PJL · 5th Jul 2018, 9:40 am

You are using full wave rectification? The trick here is to use Schottky diodes. A 1N5819 as an example has a forward voltage of <0.45V.

egerton · 5th Jul 2018, 9:49 am

Yes I am using full wave (1N4001) components I had to hand but I may change these for Schottky diodes to get more head room even though the regulator seems to be coping ok - thanks for the heads up. I may also try 1/2 wave in light of other posts here to see if that works out ok. It's fun to refresh my basic electronics. Thanks to all ideas here! Paul

egerton · 7th Jul 2018, 10:22 am

I'm now satisfied with how it works and with little impact on the original circuit. I couldn't get the 1/2 wave idea working (#22) as there was still the problem of the heater line being grounded to the sets signal ground and I didn't want to change that so I used a small 1:1 coupling transformer and it works a treat. I think it has given the set a useful technology update and I can now stream internet radio, phones and any other bluetooth source to the set simply by pressing the 'Gram' button on the set. All 3 radio bands work as before so I can still use them too. I will put a removable bluetooth symbol on the front panel glass as a talking point.

Thanks for all who helped with suggestions. Paul

Ambientnoise · 7th Jul 2018, 12:20 pm

Paul, glad you got it working, it is a good way to update a vintage set. I am curious though as to why a half wave rectifier did not work. This retains the original connection of the heater supply (one side grounded) and therefore the Bluetooth negative supply line and signal common grounded too. Perhaps a hum loop or excessive ripple caused the problems. I would have expected that “floating” the Bluetooth module off the sets ground would have caused more problems but clearly not, given the signal transformer.
I am just (technically) puzzled but great if your solution works. It does have the advantage of keeping the mains transformer balanced.

Regds

Ken

egerton · 9th Jul 2018, 11:06 am

It goes something like this:- I am deriving a new DC supply (5V) from a heater winding, and that heater winding is connected to the sets chassis and signal ground rail in the set. In isolation this is fine to power my module, but between the ground (0V) of my new supply and the ground of the set it is effectively "modulated" by the 6.3V AC of the heater supply. If I connect my oscilloscope between my 0V and the sets chassis i "see" the heater supply AC. In other words, my 0V is at the heater potential at all times. When I first connected the o/p of my module to the gram i/p of the set it blew the module - in fact I saw a spark when I connected the signal screen to ground. ouch! RIP module. Take 2 and with a tiny little coupling transformer all seem happy.

ms660 · 9th Jul 2018, 11:16 am

There's a difference between a half wave rectifier and a full wave bridge rectifier, in your receiver and with no isolation a full wave bridge rectifier will blow something, a half wave rectifier won't.

Lawrence.

egerton · 9th Jul 2018, 11:33 am

I tried it and got the same result. Perhaps with hindsight I needed to make sure I used the correct side of the heater winding for 0V? One side goes to chassis the other the heater chain. I didn't think about that as it's just A/C and so just connected a diode to one side...may have been the wrong side? Oh well I'm still living and learning.

ms660 · 9th Jul 2018, 11:38 am

Yes, rectifier connected to the wrong side from the sounds of it.

Lawrence.

egerton · 9th Jul 2018, 12:17 pm

This thread may be useful info. for anyone contemplating such a conversion. I will stick with the full wave solution as the module load varies quite a lot and getting a stable supply wasn't achieved without significant smoothing (currently 660uF) and I didn't want to use even bigger smoothing cap. I am waiting for the Schottky diodes to arrive to see how that works out, hopefully improving my DC headroom and making the regulators job at bit easier. I'd like to try this conversion on one of my AC/DC sets with a dropper. Where to derive the supply? A future project.

Ambientnoise · 9th Jul 2018, 1:30 pm

I guess I just don't understand but you have a working solution so well done. Schottky diodes should certainly help the voltage headroom.

Ken

egerton · 9th Jul 2018, 4:27 pm

Thanks Ken. Looking at the attached circuit snippet, you can see the heater winding labelled (C) connects to C on the heater feed and is shown bridged across to the sets ground. I think it's important NOT to put the diode on that side. I haven't gone back and tried it on the (A) side but it seems likely it was the problem, as it should work on 1/2 wave as you say.

3rd Jul 2018, 4:29 pm	#21
Boater Sam Banned Join Date: Dec 2013 Location: Middlewich, Cheshire, UK. & Winter in the Philippines. Posts: 3,897	Re: Ferranti A1016 bluetooth project 1:1 RS transformer. I doubt the heater winding will have any problem supplying 50mA extra, all the valve filaments together could be that far out anyway.

3rd Jul 2018, 10:51 pm	#23
egerton Pentode Join Date: May 2018 Location: Beverley, East Riding Posts: 110	Re: Ferranti A1016 bluetooth project I don't think I could get enough DC out of a 1/2 wave on the heater supply. I reckon on 6.3 x 0.45 = 2.8V (less allowing for diode fwd drop?). My full wave gives (6.3 x 1.414) - (2 x 0.7) = 7.5V and I just about get enough headroom to drive a regulator with some decent smoothing capacitor.

3rd Jul 2018, 11:16 pm	#25
McMurdo Dekatron Join Date: Jun 2009 Location: Staffordshire Moorlands, UK. Posts: 5,271	Re: Ferranti A1016 bluetooth project one of these will contain 2 line-level audio transformers. http://cpc.farnell.com/unbranded/b07...op%20isolator# The beauty is that the plastic casing can either be retained or split open as it's only held together by its sticky labels. __________________ Kevin

4th Jul 2018, 11:09 am	#26
Ambientnoise Heptode Join Date: Feb 2006 Location: Marlborough, Wiltshire, UK. Posts: 915	Re: Ferranti A1016 bluetooth project The half wave rectifier will actually give a higher peak rectified voltage (one less diode) but of course will need about 4 times the smoothing cap because of the higher ripple at 50Hz versus 100Hz. You might also consider a low dropout regulator, several good chips or diy ! Ken

4th Jul 2018, 2:16 pm	#27
egerton Pentode Join Date: May 2018 Location: Beverley, East Riding Posts: 110	Re: Ferranti A1016 bluetooth project hmmm not convinced. Anyway, I don't want to have a large smoother as it will increase switch on surge current. I already have 500 uF on the full wave to keep the rail from dipping and I'd need even more on half wave. Thanks - Paul

3rd Jul 2018, 4:41 pm	#22
Ambientnoise Heptode Join Date: Feb 2006 Location: Marlborough, Wiltshire, UK. Posts: 915	Re: Ferranti A1016 bluetooth project If you use a half wave rectifier, you will retain a common ground and not need a coupling transformer? Ken

4th Jul 2018, 2:30 pm	#28
Guest Guest Posts: n/a	Re: Ferranti A1016 bluetooth project Yes, half wave will be fine, say 6.3V RMS, that's nigh on 9V peak, less diode drop and using 1000uF of smoothing and you will have at least 8V to feed a 7805 regulator. Do put 10nF or so across the diode to stop switching noise. A quick simulation at 50mA gives a minimum voltage of just over 7V, more than enough for a low drop out 7805 type of regulator. And a whole 1/2W extra from the transformer, no problem there either.

4th Jul 2018, 3:22 pm	#29
Ambientnoise Heptode Join Date: Feb 2006 Location: Marlborough, Wiltshire, UK. Posts: 915	Re: Ferranti A1016 bluetooth project Switch-on surge will be a second or so and unlikely to cause problems. It will prob be limited by the transformer winding rather than the esr of the cap. Of course, go with whatever you are comfortable with, it’s your radio ! Ken

5th Jul 2018, 9:40 am	#30
PJL Dekatron Join Date: Sep 2005 Location: Seaford, East Sussex, UK. Posts: 5,997	Re: Ferranti A1016 bluetooth project You are using full wave rectification? The trick here is to use Schottky diodes. A 1N5819 as an example has a forward voltage of <0.45V.

5th Jul 2018, 9:49 am	#31
egerton Pentode Join Date: May 2018 Location: Beverley, East Riding Posts: 110	Re: Ferranti A1016 bluetooth project Yes I am using full wave (1N4001) components I had to hand but I may change these for Schottky diodes to get more head room even though the regulator seems to be coping ok - thanks for the heads up. I may also try 1/2 wave in light of other posts here to see if that works out ok. It's fun to refresh my basic electronics. Thanks to all ideas here! Paul

7th Jul 2018, 10:22 am	#32
egerton Pentode Join Date: May 2018 Location: Beverley, East Riding Posts: 110	Re: Ferranti A1016 bluetooth project I'm now satisfied with how it works and with little impact on the original circuit. I couldn't get the 1/2 wave idea working (#22) as there was still the problem of the heater line being grounded to the sets signal ground and I didn't want to change that so I used a small 1:1 coupling transformer and it works a treat. I think it has given the set a useful technology update and I can now stream internet radio, phones and any other bluetooth source to the set simply by pressing the 'Gram' button on the set. All 3 radio bands work as before so I can still use them too. I will put a removable bluetooth symbol on the front panel glass as a talking point. Thanks for all who helped with suggestions. Paul

7th Jul 2018, 12:20 pm	#33
Ambientnoise Heptode Join Date: Feb 2006 Location: Marlborough, Wiltshire, UK. Posts: 915	Re: Ferranti A1016 bluetooth project Paul, glad you got it working, it is a good way to update a vintage set. I am curious though as to why a half wave rectifier did not work. This retains the original connection of the heater supply (one side grounded) and therefore the Bluetooth negative supply line and signal common grounded too. Perhaps a hum loop or excessive ripple caused the problems. I would have expected that “floating” the Bluetooth module off the sets ground would have caused more problems but clearly not, given the signal transformer. I am just (technically) puzzled but great if your solution works. It does have the advantage of keeping the mains transformer balanced. Regds Ken

9th Jul 2018, 11:06 am	#34
egerton Pentode Join Date: May 2018 Location: Beverley, East Riding Posts: 110	Re: Ferranti A1016 bluetooth project It goes something like this:- I am deriving a new DC supply (5V) from a heater winding, and that heater winding is connected to the sets chassis and signal ground rail in the set. In isolation this is fine to power my module, but between the ground (0V) of my new supply and the ground of the set it is effectively "modulated" by the 6.3V AC of the heater supply. If I connect my oscilloscope between my 0V and the sets chassis i "see" the heater supply AC. In other words, my 0V is at the heater potential at all times. When I first connected the o/p of my module to the gram i/p of the set it blew the module - in fact I saw a spark when I connected the signal screen to ground. ouch! RIP module. Take 2 and with a tiny little coupling transformer all seem happy.

9th Jul 2018, 11:16 am	#35
ms660 Dekatron Join Date: Apr 2011 Location: Cornwall, UK. Posts: 13,454	Re: Ferranti A1016 bluetooth project There's a difference between a half wave rectifier and a full wave bridge rectifier, in your receiver and with no isolation a full wave bridge rectifier will blow something, a half wave rectifier won't. Lawrence.

9th Jul 2018, 11:33 am	#36
egerton Pentode Join Date: May 2018 Location: Beverley, East Riding Posts: 110	Re: Ferranti A1016 bluetooth project I tried it and got the same result. Perhaps with hindsight I needed to make sure I used the correct side of the heater winding for 0V? One side goes to chassis the other the heater chain. I didn't think about that as it's just A/C and so just connected a diode to one side...may have been the wrong side? Oh well I'm still living and learning.

9th Jul 2018, 11:38 am	#37
ms660 Dekatron Join Date: Apr 2011 Location: Cornwall, UK. Posts: 13,454	Re: Ferranti A1016 bluetooth project Yes, rectifier connected to the wrong side from the sounds of it. Lawrence.

9th Jul 2018, 12:17 pm	#38
egerton Pentode Join Date: May 2018 Location: Beverley, East Riding Posts: 110	Re: Ferranti A1016 bluetooth project This thread may be useful info. for anyone contemplating such a conversion. I will stick with the full wave solution as the module load varies quite a lot and getting a stable supply wasn't achieved without significant smoothing (currently 660uF) and I didn't want to use even bigger smoothing cap. I am waiting for the Schottky diodes to arrive to see how that works out, hopefully improving my DC headroom and making the regulators job at bit easier. I'd like to try this conversion on one of my AC/DC sets with a dropper. Where to derive the supply? A future project.

9th Jul 2018, 1:30 pm	#39
Ambientnoise Heptode Join Date: Feb 2006 Location: Marlborough, Wiltshire, UK. Posts: 915	Re: Ferranti A1016 bluetooth project I guess I just don't understand but you have a working solution so well done. Schottky diodes should certainly help the voltage headroom. Ken

9th Jul 2018, 4:27 pm	#40
egerton Pentode Join Date: May 2018 Location: Beverley, East Riding Posts: 110	Re: Ferranti A1016 bluetooth project Thanks Ken. Looking at the attached circuit snippet, you can see the heater winding labelled (C) connects to C on the heater feed and is shown bridged across to the sets ground. I think it's important NOT to put the diode on that side. I haven't gone back and tried it on the (A) side but it seems likely it was the problem, as it should work on 1/2 wave as you say.