DanSteely Posted May 16, 2013 Author Share Posted May 16, 2013 Many thanks for the all your replies and the useful information: So I guess it comes down to these options: A: I get the source files from Richard Crowley & hopefully the Chinese chaps can manufacturer the boards B: I consider transposing the layout onto Vero. I know that Boatman did this but suffered from the circuit oscillating - something I don't think he ever sorted out.. C: A new Gerber file is created by (perhaps) a member of the BR for the benefit of us all.... What I can't see, on the ComClone site, is a straight forward schematic for the circuit, which would make each certainly help. I shall continue to email Richard Crowley, who did respond to my original email of a few weeks ago, so there is hope.... I'll report back. Please let me/us know if there are any other suggested routes to move this project forward..... Thanks. Dan Link to comment Share on other sites More sharing options...
Brian Posted May 16, 2013 Share Posted May 16, 2013 As designed, it's a single sided board. It'd be worth making it double sided; getting rid of the wire links and putting a decent ground plane on the top. Link to comment Share on other sites More sharing options...
DanSteely Posted May 16, 2013 Author Share Posted May 16, 2013 Would double sided automatically mean surface mount or could 'normal' through board components still be used? Link to comment Share on other sites More sharing options...
boatman Posted May 16, 2013 Share Posted May 16, 2013 Would double sided automatically mean surface mount or could 'normal' through board components still be used? No, double sided means no wire links as the tracks on the top side replace them. The components can still be through-hole. Soldering is also significantly better as the holes are plated and the solder will run along the component legs into the hole. As Brian said it's also better to have a ground plane as that will improve noise performance significantly. I never looked at the Gerber files for Comclone so didn't realise they were missing the drill file. Maybe it's time to make a proper job of it and get a double sided PCB done. Link to comment Share on other sites More sharing options...
DanSteely Posted May 16, 2013 Author Share Posted May 16, 2013 Thanks for that clarification Boatman.. Link to comment Share on other sites More sharing options...
boatman Posted May 16, 2013 Share Posted May 16, 2013 I have just had a close look at the Comclone circuit and there is one glaring error and a couple of minor ones. The Comclone output driver circuit around U3/R21/R22/R26/R27/C18/C19/C22 is basically cr@p as it will not produce any output since both inputs of the op-amp U3B are connected together with a 33 ohm resistor! I haven't looked at the PCB layout to see if it's actually correct but I suspect not. I did try building this circuit on Veroboard several years ago and failed to make it work so gave up. I don't remember noticing this error at that time, but I didn't have access to an oscilloscope then so couldn't do any proper debugging. Link to comment Share on other sites More sharing options...
timsabre Posted May 16, 2013 Share Posted May 16, 2013 I have just had a close look at the Comclone circuit and there is one glaring error and a couple of minor ones. The Comclone output driver circuit around U3/R21/R22/R26/R27/C18/C19/C22 is basically cr@p as it will not produce any output since both inputs of the op-amp U3B are connected together with a 33 ohm resistor! I haven't looked at the PCB layout to see if it's actually correct but I suspect not. I did try building this circuit on Veroboard several years ago and failed to make it work so gave up. I don't remember noticing this error at that time, but I didn't have access to an oscilloscope then so couldn't do any proper debugging. It is correct and does work like that, I have built it.Click here to download my working circuit diagram as pdf - I think it's pretty much the same as the original. Link to comment Share on other sites More sharing options...
dbuckley Posted May 16, 2013 Share Posted May 16, 2013 I've just read that the ComClone is a copy of a Clearcom MR-102, and a quick butchers at the schematic shows a similar arrangement around the final opamp for headphone drive, albeit using 22R rather than 33R. MR-102a manual which includes schematgic here (PDF, 320KB). Link to comment Share on other sites More sharing options...
bigclive Posted May 17, 2013 Share Posted May 17, 2013 I've built a psu for less than £10 all in (using boatman's audio termination), using a £5 Chinese 2a 24v psu and a £3 enclosure, supplemented with bits from the workshop for the audio termination and salvaged connectors. Beltpacks come up on ebay for less than the component costs (I've not paid more than £40 each one), so I've purchased mine rather than built them. Headsets are from an aviation company (£20 each) with the condenser mics swapped for dynamic mics (£5 each) Can I just mention that unless you get a power supply with a genuine pedigree and testing certification then you might as well be connecting your equipment directly to the mains. The cheap Chinese power supplies (including the chassis style) often have horrifically inadequate isolation between the mains and low voltage sections, with some even winding the primary and secondary transformer windings on a single bobbin section with nothing more than a single turn of thin tape between them or worse still the wires crossing so the only separation is a thin layer of lacquer. This is especially important where people are able to directly touch metalwork referenced to the supply. Link to comment Share on other sites More sharing options...
boatman Posted May 17, 2013 Share Posted May 17, 2013 .. It is correct and does work like that, I have built it.Click here to download my working circuit diagram as pdf - I think it's pretty much the same as the original. I've just read that the ComClone is a copy of a Clearcom MR-102, and a quick butchers at the schematic shows a similar arrangement around the final opamp for headphone drive, albeit using 22R rather than 33R. MR-102a manual which includes schematgic here (PDF, 320KB). A close look at the output stages for Comclone and the MR102 show that the Comclone circuit is indeed incorrect. The positive feedback resistor R26 in the Comclone circuit should be connected to the junction of R27 & C22 and not directly to pin 7 of U3. I hope that helps anyone who is thinking of trying to make it work. http://www.mydesk.myzen.co.uk/_Useful/OutputStages.jpg Link to comment Share on other sites More sharing options...
timsabre Posted May 17, 2013 Share Posted May 17, 2013 Blimey, you're right. I'll have to take mine to bits and see if I modified the board to fix that, but I don't remember doing so. Link to comment Share on other sites More sharing options...
boatman Posted May 17, 2013 Share Posted May 17, 2013 Although the input stage of the MR102 is a good design with a bandpass filter, I think the output stage of the Tecpro BP1 is much more robust. A good compromise would be combining the two. Also, there is no need for the 2k67, 20k & 6k81 - high precision 1% resistors (read expensive) in the MR102/Comclone design, standard 2k7, 20k & 6k8 will be fine as long as they are still 1%. There are very few resistors these days which aren't 1%. http://www.mydesk.myzen.co.uk/_Useful/TecproOutput.jpg Link to comment Share on other sites More sharing options...
Brian Posted May 17, 2013 Share Posted May 17, 2013 Given that the headset should be wired as a floating load you could use a bridged amplifier chip to drive it giving much greater swing (=volume). I might take this on; I've been thinking of replacing my comms system for a while now. Either that or I need to get my packs repainted and serviced. If I did I'd go for a PCB which would fit into an extruded ali box section with all components, including XLRs, pots and switches mounted on the PCB. The design would include punched ali end panels. Link to comment Share on other sites More sharing options...
Brian Posted May 17, 2013 Share Posted May 17, 2013 So, a quick calculation over a cup of tea shows that it'd cost around £35 for a complete kit of bits to build the BlueClone. That includes... * Case - extruded anodised aluminium, can be painted if desired or bought as black anodised for an extra £5* End-panels - blank aluminium panels for you to drill or, if there was interest we could get some punched out* PCB - double-sided fibreglass board. All components board mounted for ease of construction.* XLRs - board would have holes for PCB mounted Neutrik connectors.* Switches/Pot* All components - everything would be through-hole for ease of assembly* Belt Clip Everything would be available from Rapid/CPC/Farnell. The finished unit would be about the same physical volume as a Metro/Stonewood/Granite unit but would be wider and shallower at 110mm wide, 45mm 'thick' and 83mm long. It would lack the overhang of those units to protect the knob and switches so in that respect it'd be more like a Tecpro. Link to comment Share on other sites More sharing options...
boatman Posted May 17, 2013 Share Posted May 17, 2013 Which bridge amp chip are you thinking of using? All the ones I found after a quick search are surface mount and low voltage (mostly 5v class D). 24V is also too much for TBA820 or LM386 although including a 78L12 regulator to use with either one is a possibility. I also have the full version of Easy PC (V16) if you're thinking of a co-operative venture and have PCB footprints for Neutrik NC3MAAH & NC3FAAH2 which I needed for my cuelight system. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.