Important part of that comment was the transformer current spec, which is too often ignored. Man! We gotta have fuses!!
What fuses amperage and type do you recommend for V+ and V- at the amplifier board? These fuses need to be added to the schematic.
When it comes to audio amplifier:
This is good with:
36va dual secondaries, dual bridge rectifiers
50va center tap wiring, one bridge rectifier, plus the loss of CRC's filters.
Last edited:
Maybe I've over-read something, however does the heatsink size for the LM1876 and LM3886 matter in terms of "it cannot be bigger than this by this"?
I seem to have read it in two places on the net whilst researching a first time amp project for two DIY 30W 8ohm (Visaton FRS8-8) speakers for near-field use on my desktop via laptop output - that you must make the heat sink at least a size, but also not too big. Have I got it wrong or is the "too big" a rather lose factor?
EDIT: Will it be good to ad a NE5532 pre-amp to this?
I seem to have read it in two places on the net whilst researching a first time amp project for two DIY 30W 8ohm (Visaton FRS8-8) speakers for near-field use on my desktop via laptop output - that you must make the heat sink at least a size, but also not too big. Have I got it wrong or is the "too big" a rather lose factor?
EDIT: Will it be good to ad a NE5532 pre-amp to this?
Last edited:
I didn't think that there was any such thing as a too large heatsink. Is it a myth? I've used a huge transmitter heatsink to line up a bunch of slightly different amplifier configurations, like a 12 channel LM1875, for comparison purposes. They didn't explode because of too large heatsink. None of them exploded. None were damaged.
There's something that you Can get too big, but it isn't the heatsink. . .
Actually, not exploding has to do with transformer VA. If your transformer is larger than VA*0.65=W, then you've got more transformer current than the amp can pass, in which case the extra current breaks the amp. Just like a discrete amp, with LM1875, the current limiter IS the transformer. And, a too-big transformer = a request for breakage.
SO, 36va transformers powering LM1875 monoblocs doesn't explode (one transformer, one channel = monobloc).
However, for stereo amplifier (one transformer, two channels = stereo), 64va transformer powering a pair of Parallel LM1875 amplifiers (4 chips for stereo amp) makes sense since the Parallel amp can handle that 64va.
So, at this time, I'm not worried about a too-big heatsink--I'm worried about seeing far too many too-big transformers. Especially problematic is the photos showing a melon size transformer powering a fingernail size chip amp directly followed by a question of why the amp exploded or made a lot of DC offset (caused by internal breakage). Well, the answer is that LM1875 doesn't have the SPIKE limiter--LM1875 has the same power needs as really small size discrete amplifier.
Overlarge heatsink is okay; however, overlarge transformer is Not okay.
There's something that you Can get too big, but it isn't the heatsink. . .
Actually, not exploding has to do with transformer VA. If your transformer is larger than VA*0.65=W, then you've got more transformer current than the amp can pass, in which case the extra current breaks the amp. Just like a discrete amp, with LM1875, the current limiter IS the transformer. And, a too-big transformer = a request for breakage.
SO, 36va transformers powering LM1875 monoblocs doesn't explode (one transformer, one channel = monobloc).
However, for stereo amplifier (one transformer, two channels = stereo), 64va transformer powering a pair of Parallel LM1875 amplifiers (4 chips for stereo amp) makes sense since the Parallel amp can handle that 64va.
So, at this time, I'm not worried about a too-big heatsink--I'm worried about seeing far too many too-big transformers. Especially problematic is the photos showing a melon size transformer powering a fingernail size chip amp directly followed by a question of why the amp exploded or made a lot of DC offset (caused by internal breakage). Well, the answer is that LM1875 doesn't have the SPIKE limiter--LM1875 has the same power needs as really small size discrete amplifier.
Overlarge heatsink is okay; however, overlarge transformer is Not okay.
Last edited:
Maybe too-big heatsink actually means that you shouldn't Pay for the heatsink?
Older computer CPU coolers in approximately 3.5"x3.5"x2.5" at one per each LM1875 chip, are usually enough, and those heatsinks are recycled/free. In that way, the heatsink doesn't cost anything.
P.S.
Tab has voltage. Don't forget the shoulder washers, white thermal paste and thermal pad or mica.
Older computer CPU coolers in approximately 3.5"x3.5"x2.5" at one per each LM1875 chip, are usually enough, and those heatsinks are recycled/free. In that way, the heatsink doesn't cost anything.
P.S.
Tab has voltage. Don't forget the shoulder washers, white thermal paste and thermal pad or mica.
Last edited:
Where do you get such crazy notions from?
None such thing happens.
Absolutely not, unless for you "normal" is the amplifier driving a short.
Transformer current *capacity* , above what's required, is a bonus, not a liability.
Rather than relying on a way too weak PSU as "protection"
, work on a good protection circuit, but start by properly loading the amplifier.Oh, now I see it, this is another way to push your obsession, parallel chipamps.
Melon or even watermelon size does not matter, as far as it supplies the proper voltage; extra current capacity is, as I said before, a bonus.
To avoid exploding amps (chip or discrete), start by properly building and loading them.
In that case, you might want to choose a different chip equipped with a more thorough internal limiter. LM1875 isn't well protected.
Question: Are their some ST brand replacements that have well working current limiters on-chip?
I haven't played with the smaller size chip amplifiers from ST, but I'd love to hear news about some that are well protected. Do they have any?
Question: Are their some ST brand replacements that have well working current limiters on-chip?
I haven't played with the smaller size chip amplifiers from ST, but I'd love to hear news about some that are well protected. Do they have any?
Yeah but I think I was talking about the fact that only one secondary was conducting at a time. You still get the same current amplitude. If the voltage is the same, how could you not? (Well, if your transformer VA rating was too low, you might not, if you were pushing against that limitation. But that could also be true for any other transformer/rectifier configuration.)
I seem to have read something along the line that the LM chips need some resistance - pretty much boiled down to (with my infinite electronics knowledelge ) that the chips need to reach an operating temperature. Obviously this didn't make sense but since I know so much about ths stuff thought I'd rather ask. Now I know and need to decide on a LM project or a TK2050 project (like Hifimediy T1) - plus pre-built or get the boards and get the parts and solder myself - will likely be cheaper getting a pre-assembled kit - but where is the fun in that.
Thanks.
) that the chips need to reach an operating temperature. Obviously this didn't make sense but since I know so much about ths stuff thought I'd rather ask. Now I know and need to decide on a LM project or a TK2050 project (like Hifimediy T1) - plus pre-built or get the boards and get the parts and solder myself - will likely be cheaper getting a pre-assembled kit - but where is the fun in that.Thanks.
Last edited:
I agree. Dan just makes it up as he goes along.
Usually he hides all his pet hypothesis in gobble de gook so that we don't get the chance to refute his rubbish.
Sometimes he types in plain english and then we see how bad he is with his suggestions.
John Linsley Hood, in his book, The Art of Linear Electronics, explains about the many advantages of paralleling.
Hey Daniel
I have managed to get both boards working. Third time lucky.
I also managed to get the Audio Sector 3875 kit built. Making the kit helped me to understand what I was doing wrong with the 1875 boards. I bought an extra PS with the 3875 kit. What would be the best way to connect the 2 1875 boards to it. I'm a bit confused with the 3 wires off each amp and the 8 terminals on the rectifier board.
Thanks for all your help
I have managed to get both boards working. Third time lucky
.I also managed to get the Audio Sector 3875 kit built. Making the kit helped me to understand what I was doing wrong with the 1875 boards. I bought an extra PS with the 3875 kit. What would be the best way to connect the 2 1875 boards to it. I'm a bit confused with the 3 wires off each amp and the 8 terminals on the rectifier board.
Thanks for all your help
Whoa! Dude, I think that's a bit strong ... he's not ... Mr. Burns!
(Mmwhahahaha aa, take that Internet! That will corrupt your electronics training and fry your tweeters!)
Look, I'm guessing that Dan doesn't have formal training in this type of analog design and comes at it from a hobby, DIY direction. In fact he comes at it like some sort of psychedelic ninja! He totally dives into projects and without that formal scaffold, may pick up some 'alternative' theories along the way.
Yes he may present them as general diy tool kit but as far as I can tell is open to correction. Any further nefarious motives are uncalled for.
I was very much like him although not as competent and productive. He's excited and wants to share his discoveries. He's not trying to misinform.
Last edited: