Golborne Vintage Radio

Full Version: Hedghog 625-405 standards converter
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2 3 4
I've put this item here as its directly related to vintage sets.
Frank Cuff has over the last few months been working on a standards converter. A call out of the blue a couple of weeks ago to me from Frank gave me the brilliant news that he was sending one to me. I was delighted and very honoured for the very kind gesture. Well it arrived today, I got it all set up and have been doing some subjective tests and comparison with SCRF Aurora. Now I am not going to make remarks that would upset Darryl or Frank as both converters are brilliant.
Franks converter in my eyes is brilliant, the resolution is subjectively better, it easily reproduces the finest frequency bars from 625 line test card C fed from the Imogen test card generator whereas the Aurora is much softer, watching a program reminds me of 405 lines after duplication and the excellent pictures from the BBC digital standards converter. Fast moving scenes are reproduced with no aliasing. The addition of the function switches on the front panel are an excellent idea, having to remove the cover or reprogram the Aurora is fiddly. So I have to say without treading on any toes the Hedgehog of Franks wins my approval. No disrespect to Darryl
I'll do a few more technical tests, some pictures and a video if I get Franks approval.
Many thanks to frank for sending me the converter.
I've been following the development of this converter for some time: glad to see it has come to successful fruition.

Do you have any broadcast type test equipment that you can apply to the converter to get some objective measurements?
I also have a Hedghog thanks to Frank's generosity. I've put multiburst into both an Aurora and a Hedghog. The video outputs look very similar. This isn't surprising as both use the same TVP5150 video decoder chip. This chip has a good comb filter decoder. The analogue filtering at both input and output is simple on both designs and has little effect on HF response. The Aurora has an oversampling digital filter ahead of its DAC. I don't think the Hedghog does, so in theory its analogue output filter is more critical.

I'm surprised that Trevor is seeing significant differences.

I admit I haven't looked at the modulated output - I don't have any TVs that I can sufficiently trust to reliably resolve the highest frequencies in a 405 signal.

So my initial impression is that the basic video performance of both Aurora and Hedghog is as close to ideal as it needs to be. The higher frequency bursts look a bit odd due to filtering in the decoder chip so it's hard to say that the HF response is flat within a certain figure. Pretty sure it's within 2dB up to full BW. It may well be better. What I can't comment on yet is the vertical interpolation, simply because I haven't really looked. There's nothing blatantly wrong and it's not the easiest thing to quantify.

There will be no temproal aliasing as there is no temporal processing in either Hedghog or Aurora.

When I have time to do so, I hope to do more detailed measurements.
Jeffrey I'd say there is little in it to be fair but with the "soft, medium and sharp" switch in the Medium or sharp positions there is notable improvement overall on the picture. The gratings are still visible but the higher frequency ones do appear clearer on "sharp".
I've made a small error in my first post and it's resolving the 5.25Mhz (625 line test card C, converted) they are just visible on Franks converter, the set needs to be spot on though, on the Aurora they are not visible.
I include some photo's off screen, I really need to do this with my proper digital camera on manual, at least it does give some idea.
This is test card "C" 625 lines converted by the Hedghog to 405 lines on Channel 9

This one is feeding the video from the Hedghog into a broadcast standard modulator on CH11, the fine pattern is digital noise from the Imogen test card generator.

This is the test card fed to a later Aurora SCRF and outing on CH5

This is the inbuilt PM5544 test card outing from the Hedghog on CH9

Again sorry for the rather poor pictures but as can be seen there is little difference, the Modulator on both the Hedghog & Aurora do a good job, the broadcast modulator on CH11 is very good and without the pattern from the Imogen gives the best resolution on 405 of all the other sources.
Looking carefully you can see that the result from the aurora is slightly flatter.

I will do a video over the next few days with my decent camera although it is limited to 4 minutes, should be long enough to give a reasonable test.
Looking quickly at the waveforms all seems correct but I'll look in depth tomorrow, although Jeffrey of course is the expert here.
I'm not 100% usre but I think the soft/medium/shap switch only affects vertical interpolation. It should have no effect on the frequnecy response.

I've got Frank's VHDL code here. As far as I can tell, the selection is between 2, 3 or 6 line interpolation.
Thanks for the clarification Jeffrey.
I'm convinced it is a brilliant converter, it has advantages with its front mounted switches and features.
After buying a good few auroras for myself and supplying them to customers I have to say they are reliable and easy for the one set collector to use.
I do firmly believe that now we need a good source of System A signals if the hobby is to continue and an alternative to the Aurora is really a must, none of us will be on the planet for ever.
Frank has made a wonderful piece of kit here and for me I firmly believe it has the edge on the Aurora.
Assuming Frank is willing to release full details of the Hedghog then it becomes an open source design which could, in principle, be made by anyone who wishes to do so. Small volume PCB manufacture is now inexpensive. SM assembly is feasible, if tricky, for the amateur.

One limitation is the rather old Altera Cylcone 2 FPGA. I wonder how long that little board will be around. There would be no theoretical difficulty in transferring the design to another FPGA but the physical construction would change. I think this is the FPGA board but I'm not 100% certain:
In fairness I should add that the Aurora uses an end of life Xilinx Spartan 3E FPGA. They're still available but all Spartan 3 series parts are rapidly heading for "last time buy". I'm well aware of this as I use them in a professional design. Small quantities of the more popular parts should be around on the grey market for a while but at what cost. If Darryl wants to continue making Auroras he will probably have to migrate to a newer FPGA.

Another limitation of the Hedghog is the test card. It's a preset design, generated by algorithm rather than stored data. It therefore cannot generate test card C or other arbitrary patterns.

A limitation on both the Aurora and Hedghog is the continued availability of the modualator chips. ISTR Darryl saying they were nearly obsolete.
So perhaps the original Aurora less a modulator was a good idea?
What is in the World Converter?
The thing that made the SCRF such an attractive proposition is that it's plug and play. No need to make a modulator. The SCRF is entirely usable with an external modulator but you wouldn't make and sell it without an internal modulator.

I think the WC-01 uses the same modulator chips as the SCRF. I don't know if Darryl will make any more of them as they never sold in any quantity. They weren't cheap to make and you could buy several SCRF for the price of a WC-01. The main use for a WC-01 is to make 240 line Baird and other standards that need field rate conversion.
One reason to keep the old multi standard Aurora in tip top condition.
Pages: 1 2 3 4