Hi Matttcattt,
I don't think that your transformer will be up to the job.
As I haven't built either version, I can't comment on the differences. For what it's worth, I will be building the inverted design very soon.
Sorry, what do you mean by 'explain the classes'?
Your pot should be OK but an attenuator will be needed to get the best out of this design.
For loudspeaker protection build yourself a Velleman kit. I think these are still available through Maplin.
May I humbly suggest that you take a look at my web site where there is a wealth of information for anyone beginning hi-fi DIY.
I don't think that your transformer will be up to the job.
As I haven't built either version, I can't comment on the differences. For what it's worth, I will be building the inverted design very soon.
Sorry, what do you mean by 'explain the classes'?
Your pot should be OK but an attenuator will be needed to get the best out of this design.
For loudspeaker protection build yourself a Velleman kit. I think these are still available through Maplin.
May I humbly suggest that you take a look at my web site where there is a wealth of information for anyone beginning hi-fi DIY.
I don't think an amp will be enough; especially if you have inefficient speakers. I am currently powering my gainclone with a pair of linear power supplies (one per rail), and each one deflects up to 2-3 amps per supply at full volume…. Of course, I only need full volume for the cheapo store bought speakers.. 🙂 On Brian’s Thors (87db I think), I never drew more than half an amp (for normal listening volume).Matttcattt said:
--Jordan
loudspeaker protection and muting
Mattcatt - Try this one for loudspeaker protection:
http://www.sound.westhost.com/project33.htm
Mattcatt - Try this one for loudspeaker protection:
http://www.sound.westhost.com/project33.htm
I haven't finished my gainclone for now (still waiting for the transformer)
but the 1000µF TOTAL capacitance per rail seems really low.
A few mathemetics:
ask a 1 Farad cap to deliver 1 Ampere, it will loose 1 volt per second
do this with a 1000µF one, it will loose 1000 Volt per second, in a 100Hz period (i'm in europe, 50Hz AC, 100Hz after the rectifier bridge) wich is 0.01s, it will loose 10V
this is with 1 amp, with 2.5 amp you totally discharge your cap between 2 transformer periods
but the 1000µF TOTAL capacitance per rail seems really low.
A few mathemetics:
ask a 1 Farad cap to deliver 1 Ampere, it will loose 1 volt per second
do this with a 1000µF one, it will loose 1000 Volt per second, in a 100Hz period (i'm in europe, 50Hz AC, 100Hz after the rectifier bridge) wich is 0.01s, it will loose 10V
this is with 1 amp, with 2.5 amp you totally discharge your cap between 2 transformer periods
my maths looks correct, but not good 😉
those results are everything but good
how can people obtain good sound with 1000µF; and especially better sound that with 10 000 or 50 000µF??!!
those results are everything but good
how can people obtain good sound with 1000µF; and especially better sound that with 10 000 or 50 000µF??!!
Matttcattt said:
Matttcattt,
If this is a dual 30V/1A ADJUSTABLE supply that you can set to dual 25 V at 1A I would contrary to the other responses say that this would be just perfect for the job. Put a 1000-2200 uF cap on the output of it (if it does not like it then don't) if you wish and run as many tests you want until feeling ready for the real PSU. As you clearly states this is for TESTING in which case it could almost not be better...go ahead is my recommendation.
Bricolo said:
Shame on you for muddying up an audio group with facts and figures! This is all about magic, and art, and feelings...
The "original" amp from Japan that so many people are trying to copy used 1000 uF per rail. The "designer" claims it sounds better that way, nevermind the numbers. This is a classic marketing 101 technique- make a few outrageous statements about your product and watch the debate start, all the while your name gets mentioned over and over. The very existence and length of this thread, and of other forums devoted specifically to these amps, is proof of the effectiveness of the technique.
This technique has been used most successfully and for many years by purveyors of audio cables and tweeks. This is the first instance I know of where the product is not a passive device (except maybe the CD demagnetizer, but it too is passive in the sense that it supposedly does its job on the disc and is not part of signal processing eletronics). This simple technique enables people with limited or no engineering knowledge and limited budget to create a product and sell it, especially when the target market is equally or more ignorant of the applicable engineering principles. Take a simple, inexpensive circuit, wrap it in a few hundred dollars worth of metal work, and people will line up to buy it.
I can imagine no scenario where high ripple in a power supply is going to make an amp sound "better". It may make it sound different, but adding hum and/or distortion or whatever the effect of high ripple in the supply will be, is not my idea of hi-fi.
It is a source of great personal frustration that so many people choose ignorance over education. Some even go so far as to imply that an engineering education renders one incapable of enjoying the finer points of audio reproduction. This is one of the few areas where I have seen people actually reveling in their own ignorance. The only explanation I can think of is that in the short term, ignorance is much easier than education.
Build an amp using these chips. They are well designed by well educated engineers who knew exactly what they were doing as you can clearly see by the performance detailed in the data sheets. Use plenty of capacitance in the power supply to duplicate as closely as possible the conditions under which the data sheet specs were measured. You will have a great sounding, practically indestructable (as long as you follow the design info in the data sheet and ap-notes) amp for very little cash outlay. What more could you ask for?
Now if you'll excuse me, I feel the temperature rising and I must don my fire proof suit before the flames start...
MR
massimo said:P.S. does Massimo Vimercati owe you a lot of
No just another massiomo from Milan i have cooresponded with about hifi.
dave
Bricolo,
Do as MRehorst suggests, you can go wrong following the advice of an educated engineer.
But make provision for 1000u of capacitance per rail, right beside the chip. Listen to the amp for a while and then disconnect the big caps bank and decide for yourself what you prefer. You can loose this way and you won't be dependant on your math only.😉
I did that, and I made my non educated choice. 😉
Do as MRehorst suggests, you can go wrong following the advice of an educated engineer.
But make provision for 1000u of capacitance per rail, right beside the chip. Listen to the amp for a while and then disconnect the big caps bank and decide for yourself what you prefer. You can loose this way and you won't be dependant on your math only.😉
I did that, and I made my non educated choice. 😉
Nuuk said:
why am i so blind? cool website by the way 🙂
UrSv said:
good, tankyou for the advice.
JordanG said:
i have Goodmans Mezzo SL loudspeakers (or at least thats what it says on the back), i dont know how efficient they are.
David B said:
thanks, i might be lazy and buy a kit.
another question: will 1000 - 2200uf on the supply be enough or do i need more ? i know that there are extra ones that go on the chip itself.
[edit]: i have just ordered 5 LM3875TF samples. 😀 where can i get samples of all the other components?
Matttcattt said:
Read back through the last few posts or the whole thread and make your choice. The orthodox GC followers seem to prefer 1000 uF for 18 VDC and/or 2200 uF for the higher supplies. People following other class AB recommendations like more. Some people like lots more. Others really don't know or care. I would say that this is where you will have to decide for yourself...
- Status
- Not open for further replies.