I'd buy a kit rather than built as easier to change bits if needed eg
https://www.aliexpress.com/item/100...34472037906269eaf94!12000029794591801!rec!GB!
Looks like I will be building it. Single channel. Cap's ~£23. Rather high quality as the supplier sold these in one off rather than packs. That included mica insulators, paste and one trimmer. also a 100v bridge rectifier, pence. Then a 24v 3.2anp switcher ~£16. 1500uF max on it's load. 150mm square heatsink that I think will handle one channel ~£20 off amazon. Will need to check it's thermal resistance.
A suitable transformer would cost around the same as the switcher. I'd like to try both ways to see if Nico is right. The large cap for LH's power supply did cost a couple of quid. That and the bipolar parts could be used for a fully reg'd supply.
Had problems finding some cheap plain copperless matrix board but did. The bipolar parts wont add much. Ordinary vero board could easily cause problems with high ft transistors around or swamp their effects. Prototyping boards can be just as bad
Compare this with the cost of a built stereo one on a heatsink that is highly likely to handle 10w stereo. It can be bought on ebay, locally in the UK but probably from the EU really. It's cost is less than the above heatsink.
https://www.aliexpress.com/item/100...34472037906269eaf94!12000029794591801!rec!GB!
Looks like I will be building it. Single channel. Cap's ~£23. Rather high quality as the supplier sold these in one off rather than packs. That included mica insulators, paste and one trimmer. also a 100v bridge rectifier, pence. Then a 24v 3.2anp switcher ~£16. 1500uF max on it's load. 150mm square heatsink that I think will handle one channel ~£20 off amazon. Will need to check it's thermal resistance.A suitable transformer would cost around the same as the switcher. I'd like to try both ways to see if Nico is right. The large cap for LH's power supply did cost a couple of quid. That and the bipolar parts could be used for a fully reg'd supply.
Had problems finding some cheap plain copperless matrix board but did. The bipolar parts wont add much. Ordinary vero board could easily cause problems with high ft transistors around or swamp their effects. Prototyping boards can be just as badCompare this with the cost of a built stereo one on a heatsink that is highly likely to handle 10w stereo. It can be bought on ebay, locally in the UK but probably from the EU really. It's cost is less than the above heatsink.
JLH's driver transistors, like the 2N697, 2N1613 or 2N1711, are all fairly similar TO5 or TO39 amplifier/switches. They're still available at a price but there are plenty of modern types that would equal or better them, according to what we believe are the most important features to have. Currently available versions of those transistors have Fts of about 70MHz.
BD139 by STMicro, is quoted at Ft 100MHz but there are competing manufacturers with varying specs so you may need to decide what you want most. ST's seem to work very well in the pair of amplifiers I assembled, but so did popular Chinese copies of the obsolete Hitachi 2SD669A. Watch the Hfe though, some of these general purpose semis may be ungraded and very high - perhaps unstable in some applications. I use UTC (Unisonic) product if available.
BD139 by STMicro, is quoted at Ft 100MHz but there are competing manufacturers with varying specs so you may need to decide what you want most. ST's seem to work very well in the pair of amplifiers I assembled, but so did popular Chinese copies of the obsolete Hitachi 2SD669A. Watch the Hfe though, some of these general purpose semis may be ungraded and very high - perhaps unstable in some applications. I use UTC (Unisonic) product if available.
Last edited:
This is the one with the heatsink. Kit costs £1 less than built
https://www.aliexpress.com/item/400...029167116!sea!GB!0&curPageLogUid=w70RNPyYoLW5
Actually that one wouldn't be too bad for trying different transistors, Snip the leads and fit pins,wire etc. Snipping makes it easier to desolder without damaging the board. Repeated resoldering often does. Solder suckers often work more effectively if more solder is added first.
Cost differences. Part of that will be down to the nature of the electronics industry some down to cheaper parts but a lot down to the industry. The major retailer Mouser etc have a pretty easy life and the actual part manufacturers want ever increasing order sizes. Even certain logiistic companies have moved into it.
Kit costs £1 less than builthttps://www.aliexpress.com/item/400...029167116!sea!GB!0&curPageLogUid=w70RNPyYoLW5
Actually that one wouldn't be too bad for trying different transistors, Snip the leads and fit pins,wire etc. Snipping makes it easier to desolder without damaging the board. Repeated resoldering often does. Solder suckers often work more effectively if more solder is added first.
Cost differences. Part of that will be down to the nature of the electronics industry some down to cheaper parts but a lot down to the industry. The major retailer Mouser etc have a pretty easy life and the actual part manufacturers want ever increasing order sizes. Even certain logiistic companies have moved into it.
Did you try Farnell? - their "multicomp" product isn't large, but cheap enough. https://www.farnell.com/datasheets/2313197.pdf
Last edited:
I found sone at RSWW, epoxy paper 160mm long. No signs on CPC which is Farnell's more consumer outlet. It may be on ebay. Didn't look. Suppose people more often want pads or tracks.
It's more a longer term project for me as main interest is speakers and my USB mic should arrive this week. The amp will do for a higher quality test amp than the one I have which is a small bare board prebuilt digital one. The LH amp has always fascinated me as well. I have heard it in action rather quickly knocked up. It's way way more easy to thoroughly test amps these days than then. I don't think the change in sound with it has much to do with distortion levels. More the amount of current around to control the loudspeaker and lack of crossover. Single ended chat I am afraid makes me laugh, Clearly the top tr could be replaced with a current source. It may get around hfe matching. Try it and find out. I wonder why LH didn't or maybe he did.
Most of the money I have spent to be able to build one from scratch can also be used for other things including snipping the bipolar parts off the Chinese built stereo amp I linked too. That one is also fully adjustable. Be interesting to see if one of the trimmers melts.
Resistors bug me a bit. Automotive background. The usual 1/4w resistor kit would be rated 1/8w there. I wonder if there has been a change in this area since 1969. I haven't built much for a long time and the ratings caught me out once. Pass the experience just seemed odd and not experienced many years ago. Too many years ago.
What would you hope to achieve or improve by using massive or a large number of polypropylene film caps for the output capacitor? I think you'll find there are easier ways to test such ideas, the simplest being to add perhaps just 1-5 μF in parallel with the existing electrolytic type (that should already be a high quality type) and I doubt you need MKP type at loudspeaker impedance level and frequency. Much cheaper MKT (polyester, mylar etc.) type film capacitors should achieve all you could expect from using film caps for the audio output.
As you already have a large 160uF MKP type, if it could be mounted securely there's no harm in trying it in parallel with the electrolytic cap. in a conventional loudspeaker arrangement too. Please, let's "hear" about it!
As you already have a large 160uF MKP type, if it could be mounted securely there's no harm in trying it in parallel with the electrolytic cap. in a conventional loudspeaker arrangement too. Please, let's "hear" about it!
Last edited:
I believe automotive applications focus on reliability under physical duress like vibration and tolerance to abnormally high temperature. That's likely cheapest by just upping the power rating spec. a grade or two. On the other hand, plain film resistors that DIY folk purchase now, usually come from the orient, are extremely cheap but unless you've been unwise, they won't be facing the same physical demands or cost reduction measures.
Over a decade or so, consumer electronics used to be plagued by insurance requirements to fit original spec. safety (fusible) resistors in many applications. That was and remains a real PITA for obtaining suitable replacements. So you write off the whole appliance because you can't get the right spec. 10 cent components any more!
Hi. I asked about can I use only film capacitor and use it as a high pass filter to drive Tweeter or midrange. Is it ok not to use electrolytic polar capacitor and only use low value film capacitors for biamping as dc block capacitor.So we may not use capacitors 2 times on tweeter this way
This is probably best replied to in the loudspeaker forums. Since you may be looking to do what is best done with active crossovers, it's better to have a small crossover network to precisely divide the HF and LF frequencies at low signal level and let the bass and treble speakers be driven by separate, high and low power amplifiers (i.e. the bi-amps.) JLH amps are often used to drive tweeters alone and you don't need to spend much at all on the caps.
Last edited:
The design spec is -40C to +85. Industrial is likely to use the same. Military even higher. Commercial is 0 to 70C but Japan have used -10 to 70C in the past. So designers work to those limits. Similarly manufacturers specify max junction and part temperatures etc They don't invent them. They are meaningful. Data given will be temperature related, also meaningful.
In my case over time resistors could have come from a high street shop, RSWW or Farnell. Probably in 10's from the last 2. Buy a kit of so called 1/4w and run into a problem. Mouser etc are primarily industrial suppliers. The 2 similar I mentioned clearly supply 1/4watt as I would expect also the shop. The max temperature of a common resistor is usually ~150C. Few would choose to run a 1/4w at 1/4 and similarly 1/8w at 1/8w. However when some one chooses to run a so called 1/4 at 1/8w whoops as it's really an 1/8w. A real 1/4 would get hot anyway.
Design data is available for resistors eg
https://www.vishay.com/docs/31019/ptf.pdf
So I wonder where Mr Hood bought his bits from.
Reply to #9211
By the way, you can add a tweeter to any loudspeaker that has poor or no high frequency response by just adding a capacitor for a 1st order, 3dB/octave filter and a matched sensitivity tweeter. However there are usually significant conflicts between the relative sensitivity, phase and roll-off characteristics of different types of speakers Their summed response can often make the audio seem peaky, dull sounding or even resonant, regarding the total acoustic output. The final result tends to sound like cheap, after-market products though, where the only similarity to original products, may be their appearance.
By the way, you can add a tweeter to any loudspeaker that has poor or no high frequency response by just adding a capacitor for a 1st order, 3dB/octave filter and a matched sensitivity tweeter. However there are usually significant conflicts between the relative sensitivity, phase and roll-off characteristics of different types of speakers Their summed response can often make the audio seem peaky, dull sounding or even resonant, regarding the total acoustic output. The final result tends to sound like cheap, after-market products though, where the only similarity to original products, may be their appearance.
Last edited:
His employer, of course
.I actually worked for a UK based resistor manufacturer in my last serious job but their main process line here was outdated and too expensive on labour costs. Faced with the demands of colour TV, VCRs etc, they had sold their plant and equipment offshore by the time I had arrived and restored the laboratory to support what lines remained.
Last edited: