Practical Electronics. Battery Valve Radio Power Supply.

[Philips210]

Hi.

Just picked up January 2022 issue of Practical Electronics magazine from Smiths and noticed an interesting project that might interest you. It's a battery valve power supply using Li-ion batteries. The circuit will provide the valve heater supply as well as the HT.
Quite a useful circuit where no mains supply is required, ie it is truely portable.

Regards,
Symon

[Electronpusher0]

This is exactly the same project published by Silicon chip magazine in the December 2020 issue.

Peter

[Electronpusher0]

Originally Posted by Electronpusher0

This is exactly the same project published by Silicon chip magazine in the December 2020 issue.

Peter

The article first page is here and an electronic copy of the magazine can be purchased for 10 Au dollars here

https://www.siliconchip.com.au/Issue/2020/December

Peter

[Terry_VK5TM]

Cheaper to buy PE.

[kalee20]

So , tempt me... (nearest PE will be in Barnstaple, 20 miles away): Is there an efficiency figure quoted? And are LT and HT both derived by switching techniques or is the LT just a linear regulator straight from the battery?

[Bazz4CQJ]

Is it using four 18650 cells? Seems a lot.

B

[Electronpusher0]

Originally Posted by kalee20

So , tempt me... (nearest PE will be in Barnstaple, 20 miles away): Is there an efficiency figure quoted? And are LT and HT both derived by switching techniques or is the LT just a linear regulator straight from the battery?

Reading the first page on the link I posted above it looks like LT is derived from a linear regulator from the battery.
The HT uses a normal mains transformer "I chose a 5W PCB mounting mains transformer which I used backwards" so the switching frequency must be between 50 and a few hundred hertz.

Peter

[Radio Wrangler]

Originally Posted by Electronpusher0

so the switching frequency must be between 50 and a few hundred hertz.

That's not a bad choice for the job. Most modern switcher designs with ferrite transformers would likely be switching within the long/medium wave bands, or have fat harmonics in them, creating hell.

David

[space charge]

i prefer the ancient method of using a transformer of wwhich there are plent of designs. why build a sophisticated noise machine when old systems work well EI7KA

[Philips210]

Hi.

The LT supply is derived via a linear voltage regulator whereas the HT supply is switch mode with a typical center frequency of 72kHz.
It seems to use readily available components and no special wound parts, just a standard iron core mains transformer


To quote a few points from the Features and Specifications:

Runs from two or four Li-ion, LiPo or LiFePo4 batteries
HT output 24 to 135V DC at 2W
LT output 1.2 to 2.5V at up to 3A (with a heatsink)
Low EMI. Variable drive frequency and duty cycle.
Battery over discharge protection

Regards,
Symon

[Bazz4CQJ]

Some time ago, Joebog1 pointed out the availability of some ready-made DC-DC modules 8-32 V in, 45-390 V out, online for £1-74 (they've gone up to £2.03 now) https://www.aliexpress.com/item/3295...4d1f4c4dv69dFH

Not sure how much time and money the PE projects will absorb .

B

[Dekatron]

There's a really nice article for a 90V battery replacement from Ronald Dekker here: https://www.dos4ever.com/battery/battery.html with full explanation of the circuit and at the end he shows a quite simple but very nice design.

I haven't compared Ronald's design to the article but I have experimented with Ronald's design and found it very useful.

[kellys_eye]

Unless the PSU can fit where the original battery did I find these types to be of no more use than any transformer-derived supply kept in a box.

[Terry_VK5TM]

The PE/Siliconchip design also uses some esoteric bits as opposed to more garden variety components and is not small - as in no chance of fitting in a battery space.

There are better designs out there.

Re the 18650 cells - the LT side has two cells in parallel (design flaw no1) and HT side uses two in series.

[kalee20]

Originally Posted by Radio Wrangler

That's not a bad choice for the job. Most modern switcher designs with ferrite transformers would likely be switching within the long/medium wave bands, or have fat harmonics in them, creating hell.

I can vouch for that - the interference is the hardest nut to crack!

I made a converter this year providing HT, LT, and GB- operating from a 12V SLA battery. Overall efficiency is around 80% (it's based on a flyback converter running at 25kHz), but I needed 3-stage LC filtering on the HT output, the LT output, and the 12V input (and 3-stage RC filtering on the GB output) to keep it quiet.

Working at low frequencies does offer the opportunity to use slow switching devices, keeping RFI right down, while not hurting efficiency by much because there are fewer switching events per second. The down side is that the size of the transformer increases. But, it's still not going to be massive, at a few watts only if power, and I can see the attraction of using an off-the-shelf transformer.

[Bazz4CQJ]

Originally Posted by kalee20

I can vouch for that - the interference is the hardest nut to crack! I made a converter this year providing HT, LT, and GB- operating from a 12V SLA battery. Overall efficiency is around 80% (it's based on a flyback converter running at 25kHz), but I needed 3-stage LC filtering on the HT output, the LT output, and the 12V input (and 3-stage RC filtering on the GB output) to keep it quiet.

I wonder if you would like to post some details of the filtering circuits; schematics or pictures? The Aliexpress modules I mentioned earlier will need that sort of additional circuitry and I'm not very sure how best to do it.
Thanks
B

the LT side has two cells in parallel (design flaw no1)

. I ask why?

[Terry_VK5TM]

Because when used in parallel packs, either they have circuitry to prevent one or other of them being over discharged or they are very carefully matched.

There is no such provision for or mention of this in the article.

[kellys_eye]

Paralleling cells is common in laptop battery packs.

[kalee20]

Originally Posted by Bazz4CQJ

I wonder if you would like to post some details of the filtering circuits; schematics or pictures?

Here. (I have these photos on my mobile phone). The converter unit is on the right, and I added the filtering later as a 'motherboard'. It's one of those breadboards which evolved from a duplicate build of a flyback converter using 555 timers to help debug a circuit a friend was developing on the other side of the country. It then got enhanced to supply power to a radio I was developing myself, and which needed +1.2V 0.5A LT; +120V 20mA HT; and -70V GB (really!)

Circuit diagrams might follow later!

It's not at all an optimum filter, use of 25V capacitors on the 1.2V LT line is vast overkill, but they were to hand...

The chokes are hand-wound on little toroidal cores. You can see a few.

The whole thing is wrapped in bubble wrap and Bacofoil (hence the foil grounding lead). It's a lab demo setup, but it does work well.