Hi L,
stop telling fibs, we know you know.
I often read that a reactive speaker can demand more current than the equivalent nominal resistor.
An 8ohm speaker that has a DC resistance of 6r0 would expect to pass a maximum of 6Apk when 24Vpk is applied. But, this does not seem to be a good model (using Re as peak current limiter) for fast transient currents into reactive speakers.
A doubling of expected current demand is generally agreed and many report three times as much current than what an 8ohm resistor would draw.
Fairly recently one of our magazines reported on peak transient currents using real speakers fed with real music.
The team chose three different speakers, one of severe reactance and a reputation for being difficult to drive, a more moderate speaker and a third was specifically chosen for it's reputation to have lowish reactance and be an easy load to drive.
They reported that all three speakers could in exceptional (rare) cases of real music signals draw currents that were greater than 5times what a nominal resistor would draw. They also pointed out that the peaks currents were not necessarily from the same instant in the music waveform for the three speaker types.
This report is linked from a post in this Forum. I can't remember where.
On that basis, the 3times that I had adopted as my design criteria does not seem to be too extreme. I decided to stay with 3times after reading the report, even though the team convinced me that such extreme currents can be expected with real music.
part of the reason I retained 3times was that I do not often use the full power (near clipping) of my discrete amplifiers. The rare occurence of extreme transients combined with the rare occurrence of requiring full power are very unlikely to occur in my listening.
That is also the reason I like to size my amps for +20dB above my normal maximum average listening level.
stop telling fibs, we know you know.
I often read that a reactive speaker can demand more current than the equivalent nominal resistor.
An 8ohm speaker that has a DC resistance of 6r0 would expect to pass a maximum of 6Apk when 24Vpk is applied. But, this does not seem to be a good model (using Re as peak current limiter) for fast transient currents into reactive speakers.
A doubling of expected current demand is generally agreed and many report three times as much current than what an 8ohm resistor would draw.
Fairly recently one of our magazines reported on peak transient currents using real speakers fed with real music.
The team chose three different speakers, one of severe reactance and a reputation for being difficult to drive, a more moderate speaker and a third was specifically chosen for it's reputation to have lowish reactance and be an easy load to drive.
They reported that all three speakers could in exceptional (rare) cases of real music signals draw currents that were greater than 5times what a nominal resistor would draw. They also pointed out that the peaks currents were not necessarily from the same instant in the music waveform for the three speaker types.
This report is linked from a post in this Forum. I can't remember where.
On that basis, the 3times that I had adopted as my design criteria does not seem to be too extreme. I decided to stay with 3times after reading the report, even though the team convinced me that such extreme currents can be expected with real music.
part of the reason I retained 3times was that I do not often use the full power (near clipping) of my discrete amplifiers. The rare occurence of extreme transients combined with the rare occurrence of requiring full power are very unlikely to occur in my listening.
That is also the reason I like to size my amps for +20dB above my normal maximum average listening level.
Last edited:
Surely though with a protection circuit such as this, if you are going to take into account 3 times the possible current delivery value with the argument that you don't use the amplifier anywhere near it's limits, then you don't need a protection circuit at all as you are ensuring the amp doesn't overload anyway by manual intervention?
You may as well design for bullet proof, and the protection circuitry will then only activate if you start nearing the limits of the amp.
You may as well design for bullet proof, and the protection circuitry will then only activate if you start nearing the limits of the amp.
My view of protection is to protect from abuse.
I define good protection as allowing all valid audio signals to pass to all valid audio loads.
This obviously excludes fitting 2ohm speakers instead of 4ohm when the amp is specified as suitable for driving 4ohm and higher impedance.
It also excludes a momentary output short circuit.
It excludes a momentary but significant DC presented to the input.
But, and this is where protection gets a bad reputation, good protection must not interfere with a valid signal !!!!!!
If that valid signal includes very short duration transient peak currents approaching 3times the nominal maximum peak current into the rated nominal load then the protection must be designed to allow that transient to pass.
This is not bulletproofing.
If the duration (of that extreme, but valid transient) is longer than a valid audio signal then the protection must be designed to become active to prevent damage.
I define good protection as allowing all valid audio signals to pass to all valid audio loads.
This obviously excludes fitting 2ohm speakers instead of 4ohm when the amp is specified as suitable for driving 4ohm and higher impedance.
It also excludes a momentary output short circuit.
It excludes a momentary but significant DC presented to the input.
But, and this is where protection gets a bad reputation, good protection must not interfere with a valid signal !!!!!!
If that valid signal includes very short duration transient peak currents approaching 3times the nominal maximum peak current into the rated nominal load then the protection must be designed to allow that transient to pass.
This is not bulletproofing.
If the duration (of that extreme, but valid transient) is longer than a valid audio signal then the protection must be designed to become active to prevent damage.
Last edited:
Hi,
this is my second attempt to build a Leach amp. I got second-hand to-3 pre-drilled heat-sinks and ordered a 3mm plate of aluminum for enclosure.
Transformer 350VA 2x45VAC and 6x4700uf per rail - I think at 4 ohm load amplifier will be unable to delivery full rated power but its not so important to me.
There Still remaining body polish works and one channel is not completed due missing transistors. I hope that I will build another well-sounding Leach amp.
this is my second attempt to build a Leach amp. I got second-hand to-3 pre-drilled heat-sinks and ordered a 3mm plate of aluminum for enclosure.
Transformer 350VA 2x45VAC and 6x4700uf per rail - I think at 4 ohm load amplifier will be unable to delivery full rated power but its not so important to me.
There Still remaining body polish works and one channel is not completed due missing transistors. I hope that I will build another well-sounding Leach amp.
this does not matter.
The Leach using 40Vac transformer is ~150W into 8r0 and ~250W into 4r0.
A two channel amplifier would have a total maximum output of approximately 500W into a pair of 4ohm speakers.
The recommended transformer size is from 500VA to 1000VA. All transformers in this range will work well.
A slightly smaller transformer rated at 350VA will still work.
You may be able to hear a slight difference if you compare the 350VA to a 750VA, but you would have to build both stereo amplifiers to demonstrate that difference and then decide which one to sell to try to get your money back or put it down to a "learning experience".
BTW,
a 2pair output stage Leach using 45Vac transformer cannot reliably be rated for use with 4ohm speakers. 8ohms is perfectly OK, 6ohms will take the Leach to very high SOAR when driven hard.
6x4700uF happens to suit 6ohms very well, if you want wideband audio signal reproduction.
Last edited:
on 220Vac into a 45Vac transformer driving two channels of 8ohms speaker, there is no problem with selecting a 350VA instead of designing for 4ohm loading.
on 240Vac into a 230:40Vac transformer I get +-58.5Vdc after biasing the two amplifiers to 220+220mA.
+-57Vdc from a 45Vac transformer seems exceptionally low.
have you measured this?
on 240Vac into a 230:40Vac transformer I get +-58.5Vdc after biasing the two amplifiers to 220+220mA.
+-57Vdc from a 45Vac transformer seems exceptionally low.
Last edited:
If I were a transformer manufacturer, I would advertise my product as designed to be suitable for all UK domestic voltages upto the maximum of 254Vac and tell all potential customers that is why "my" product is the correct price for the correct specification.
I would probably go bust (bankrupt) and that is why I never went into sales. I have scruples.
I would probably go bust (bankrupt) and that is why I never went into sales. I have scruples.
the only way to do that is if you make a power traffo with multiple taps in the primary to accommodate various line voltages...
Yes, some of my equipment has multi-tapped primaries.
Or, simply design for worst case high voltage and accept that there is a fixed maximum current capability, even when the voltage is lower than maximum.
It really depends on the heatsinking, and how reactive the load is.
The original Crown DC300 only had two pair and was bad news. Crown went to four pair with the DC300A.
It also depends on what you are willing to accept for a life expectancy. I am seeing failures in the three pair (TO-3) Leach on ±63V at around 25 years, I think that is acceptable.
The Mitsubishi DA15-DC had two pair with good heatsinking, and I had to repair every one we sold during the warranty period. I don't think this was acceptable.
The original Crown DC300 only had two pair and was bad news. Crown went to four pair with the DC300A.
It also depends on what you are willing to accept for a life expectancy. I am seeing failures in the three pair (TO-3) Leach on ±63V at around 25 years, I think that is acceptable.
The Mitsubishi DA15-DC had two pair with good heatsinking, and I had to repair every one we sold during the warranty period. I don't think this was acceptable.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.