Hi guys.
I'm a newbie in this forum, so forgive me if my questions seems absurd and also my english which is not my native language. I studied electronics a while ago, well a decade ago. I built then a 120wRMS on 4ohms power amp using the famous hitachi 2sk1058/2sj162.
This amplifier had a tube preamp and after months of good sound it started to oscillate, well, I could'nt find then what was wrong, so I decided to unmont all pcb's, transformers etc, until I could find time and/or better knowledge, I suspected the preamp, later I discovered It was a poor heatsink/mica construction, there was no kepton available 10 years ago in the electronics shop, so the mica did a short which virtually destroyed the mosfets and of course the drivers, but that's not the point here.
Well, my transformer was 40-0-40 250VA, big heatsink, etc, everything according to calculations (+/-56VDC rails) in an old but effective design which used the even oldest hitachi lateral mosfets which came on TO-3 packages (I forgot their names, but then hitachi referred me to the 1058/162 replacements. It worked for several months, sounded great, but the oscillation (also called motorboating when referring to tubes) started and It's been in a box for several years.
About two months ago I decided to work again on the amp, and found that there are other interesting designs for the output stage, for example the P101 which I ordered the PCB and all the components. But something stopped me on my tracks... the power output for this design gives extreme power under my 4 ohms cabinet, according to Rod, the bias of the mosfets has nothing to do with the power output, the DC rails are responsible for that. (I only use 2 mosfets)
I'm looking for my old schematic on the output stage. The p101 gives 150w in 8ohms and I expect it gives around 200w give or take in a 4 ohms load with the same DC rails that my old schematics wich gave 80w in 8 ohms and around 120 on 4ohms (my cabinet is 4ohms rated 120-130w RMS). Maybe I'm a bit lazy about this, I have many books to read and a lot of theoretical questions, sure you folks have a better understanding of what I'm asking which is:
How come that with the same pair of mosfet for the output stage and the same DC rails my old amp gave 120w and the p101 on the other hand doubles this power rating? This is driving me nuts, I can't understand it, the schematic I built several years ago had very detailed information on the power ratings for different DC rails configuration, maybe I need to get back to my microelectronic books from university and read some more I have around, specific on audio pre and power amp stages.
What do you think?
I'm a newbie in this forum, so forgive me if my questions seems absurd and also my english which is not my native language. I studied electronics a while ago, well a decade ago. I built then a 120wRMS on 4ohms power amp using the famous hitachi 2sk1058/2sj162.
This amplifier had a tube preamp and after months of good sound it started to oscillate, well, I could'nt find then what was wrong, so I decided to unmont all pcb's, transformers etc, until I could find time and/or better knowledge, I suspected the preamp, later I discovered It was a poor heatsink/mica construction, there was no kepton available 10 years ago in the electronics shop, so the mica did a short which virtually destroyed the mosfets and of course the drivers, but that's not the point here.
Well, my transformer was 40-0-40 250VA, big heatsink, etc, everything according to calculations (+/-56VDC rails) in an old but effective design which used the even oldest hitachi lateral mosfets which came on TO-3 packages (I forgot their names, but then hitachi referred me to the 1058/162 replacements. It worked for several months, sounded great, but the oscillation (also called motorboating when referring to tubes) started and It's been in a box for several years.
About two months ago I decided to work again on the amp, and found that there are other interesting designs for the output stage, for example the P101 which I ordered the PCB and all the components. But something stopped me on my tracks... the power output for this design gives extreme power under my 4 ohms cabinet, according to Rod, the bias of the mosfets has nothing to do with the power output, the DC rails are responsible for that. (I only use 2 mosfets)
I'm looking for my old schematic on the output stage. The p101 gives 150w in 8ohms and I expect it gives around 200w give or take in a 4 ohms load with the same DC rails that my old schematics wich gave 80w in 8 ohms and around 120 on 4ohms (my cabinet is 4ohms rated 120-130w RMS). Maybe I'm a bit lazy about this, I have many books to read and a lot of theoretical questions, sure you folks have a better understanding of what I'm asking which is:
How come that with the same pair of mosfet for the output stage and the same DC rails my old amp gave 120w and the p101 on the other hand doubles this power rating? This is driving me nuts, I can't understand it, the schematic I built several years ago had very detailed information on the power ratings for different DC rails configuration, maybe I need to get back to my microelectronic books from university and read some more I have around, specific on audio pre and power amp stages.
What do you think?
Hi,
With 56V rails the p101 claims 150W 8ohms short term, whatever that means.
Claimed RMS figures are 100W into 8 ohms and "about" 150W into 4 ohms.
Output power for MOSFET amplifiers is heavily dependent on the Vds vs.
output current curves of the devices, that may explain the disparity.
rgds, sreten.
With 56V rails the p101 claims 150W 8ohms short term, whatever that means.
Claimed RMS figures are 100W into 8 ohms and "about" 150W into 4 ohms.
Output power for MOSFET amplifiers is heavily dependent on the Vds vs.
output current curves of the devices, that may explain the disparity.
rgds, sreten.
A single pair of Laterals of the original or later flat-pak type are really only suitable for around 70 watts/8ohm
A 100watts rms into 8 ohms means an output voltage of around 28 volts AC rms from the amplifier. That works out at -/+40 volts peak to peak. Assuming the laterals saturate at around 12 volts worst case, you need rails of -/+52 volts or more... all too much for a single pair.
A 100watts rms into 8 ohms means an output voltage of around 28 volts AC rms from the amplifier. That works out at -/+40 volts peak to peak. Assuming the laterals saturate at around 12 volts worst case, you need rails of -/+52 volts or more... all too much for a single pair.
That's one of my concerns, I want to talk about RMS power (continuous) not peaks or "musical" power, I think these terms confuse people, me included, my transformer is 250VA.
I did some math this morning following a thread in this forum. My objective is to not damage the speakers, so for a pure resistive load (which is not) of 4 ohms, 120w using ohms law (maybe I'm mistaken here using a DC power equation) P=V^2/R so 120= v^2/4 which solving for "v" i got v=21.9V and solving i=5,4A, around 118W.
Assuming a 60% of efficiency for a A/B class amp, needs a 200W power supply, which can be achieved with mine which is 250VA.
I don't think these equations should be used for RMS at all, since it's pure DC equation and so it's a peak value which is not what I'm after.
The documents I have in the old design stated 100w RMS with 40-0-40 on 8 ohms resistive load, If 150w on 8 ohms is a peak value (DC) as stated, then clearly I'm doing this the wrong way.
I will have to read more about peak power and RMS power.
One possibility which crossed my mind this morning is mount the p101, buy a dozen or so of 10w resistors and give it a try with the dummy load wich is not 100% accurate but close, and with the oscilloscope I can check the output wave and estimate the real power that the mosfets are giving just before clipping.
Thanks for the reply.
I did some math this morning following a thread in this forum. My objective is to not damage the speakers, so for a pure resistive load (which is not) of 4 ohms, 120w using ohms law (maybe I'm mistaken here using a DC power equation) P=V^2/R so 120= v^2/4 which solving for "v" i got v=21.9V and solving i=5,4A, around 118W.
Assuming a 60% of efficiency for a A/B class amp, needs a 200W power supply, which can be achieved with mine which is 250VA.
I don't think these equations should be used for RMS at all, since it's pure DC equation and so it's a peak value which is not what I'm after.
The documents I have in the old design stated 100w RMS with 40-0-40 on 8 ohms resistive load, If 150w on 8 ohms is a peak value (DC) as stated, then clearly I'm doing this the wrong way.
I will have to read more about peak power and RMS power.
One possibility which crossed my mind this morning is mount the p101, buy a dozen or so of 10w resistors and give it a try with the dummy load wich is not 100% accurate but close, and with the oscilloscope I can check the output wave and estimate the real power that the mosfets are giving just before clipping.
Thanks for the reply.
Thanks, this makes sense with the documentation of my old design, exactly the same figures.
I will mount a dummy load with 10w resistors (at least a dozen, maybe more) and check with the oscilloscope using a sinusoidal wave at 1Khz once it's working properly, because I rememberl this is the way we checked power output back in the late 90's university labs.
Hi,
Your looking at it the wrong way. Tweeters in speakers are usually blown
by too small an amplifier. The other drivers are at risk with continuous
test tones causing them to exceed their thermal limits and the bass
units at risk with simply too loud heavy bass causing mechanical
damage and/or overheating of popping (from the gap) voice coils.
With music program clipping mildly on peaks, (i.e. as loud as gets
as long as the speakers can handle the bass), average power into
the speakers is far less than the amplifiers RMS power rating.
Music spends 80% of the time at less than 20% of maximum output.
You can't prevent damage by matching speaker and amplifier power.
You can use an amplifier with enough output to blow the speakers
if they are abused quite safely if you are sensible with the volume
control depending n the type of music program being played.
e.g. String quartets can be taken to clipping level on a very big amplifier
with small speakers, but that is not a good idea at all with house music.
rgds, sreten.
The speakers are not for hifi, so no tweeter, crosssover or anything like that. Two 12" (8 ohms each, 4 in parallel). In theory the speakers should exceed the rms power of the amp because rms is sinusoidal, a square wave gives a lot more power than a sinusoidal signal, so I was taught to mismatch the power, but in the opposite way, the speakers always have to be rated bigger than the amp (I've blown tweeters because of this, the amp gave 150w + 150 w rms, the owner claimed the speakers were rated 200w, the day after, I noticed the tweeters weren't working, I opened them and saw the real rms value (80), so I had to replace those tweeters).
I do know that keeping in mind (and ears) volume is a good way of knowing if you're driving the speakers too much, but I don't want to take that chance. I'll keep reading. Thanks.
I do know that keeping in mind (and ears) volume is a good way of knowing if you're driving the speakers too much, but I don't want to take that chance. I'll keep reading. Thanks.
More maths
Did some more maths, lets see what you think:
120W over 4 ohms (DC peak claculation)
P=V.I
I=V/r
P=V^2/R
120= v^2/4
480 = v^2
v=21.9V
i=5,4A
OK, let's see these value son RMS:
VRMS = 0.7 × Vpeak
Vpeak(max)= 22V (taken from DC peak point)
Vrms = 15Vrms
Irms=15/4 =3,75Arms
Prms= 82,5Wrms
Idc =Irms.sqrt(2) = 5,4A
Someone also said that these lateral mosfets were to give arount 70w in 8 ohms, it's well in the figure, 70w on 8 ohms and 82w in 4 ohms is a good match, it kind of give the maths and replies meaning.
So we are narrowing down, that peak we wanted to achieve is really 82,5w RMS... on 4 ohm pretty lower figure taken to the claimed 220w (discovered as peak, not real rms )
Las thoughts: We are building based on a 120w rated speaker and giving that peak, transformed in 82,5 RMS, if designed so we have a peak"dangerous" for the speakers: 220W, well that was dv, speakers are rated rms, so wuite sure my 120w rms rated cabinet can handle 82,5w RMS, I must say I'm happy with the idea, building another unit I can easily have a stereo unit.
I¡m still to prove that these mosfets can handle that and comes from a reply saying the Vds of the lateral mosfets being a lower number, 12. Exicon makes a very good one lateral mosfets in TO-3 packages, rated 250V and. VDsatmax being 12V, well, question number n for today:
Is this Vds(st) a limiting factor on the design?
We can't have more than 12V on Vds before it satturate?
Did some more maths, lets see what you think:
120W over 4 ohms (DC peak claculation)
P=V.I
I=V/r
P=V^2/R
120= v^2/4
480 = v^2
v=21.9V
i=5,4A
OK, let's see these value son RMS:
VRMS = 0.7 × Vpeak
Vpeak(max)= 22V (taken from DC peak point)
Vrms = 15Vrms
Irms=15/4 =3,75Arms
Prms= 82,5Wrms
Idc =Irms.sqrt(2) = 5,4A
Someone also said that these lateral mosfets were to give arount 70w in 8 ohms, it's well in the figure, 70w on 8 ohms and 82w in 4 ohms is a good match, it kind of give the maths and replies meaning.
So we are narrowing down, that peak we wanted to achieve is really 82,5w RMS... on 4 ohm pretty lower figure taken to the claimed 220w (discovered as peak, not real rms )
Las thoughts: We are building based on a 120w rated speaker and giving that peak, transformed in 82,5 RMS, if designed so we have a peak"dangerous" for the speakers: 220W, well that was dv, speakers are rated rms, so wuite sure my 120w rms rated cabinet can handle 82,5w RMS, I must say I'm happy with the idea, building another unit I can easily have a stereo unit.
I¡m still to prove that these mosfets can handle that and comes from a reply saying the Vds of the lateral mosfets being a lower number, 12. Exicon makes a very good one lateral mosfets in TO-3 packages, rated 250V and. VDsatmax being 12V, well, question number n for today:
Is this Vds(st) a limiting factor on the design?
We can't have more than 12V on Vds before it satturate?
Yes, that's why I'm posting a thread and asking you guys.
I don't have absolute truth about anything, I don't know.
I do think that the dummy load and testing is the way to go.
Am I wrong about this too?
I'm building the P101 (high power version) to go bi-wiring with my hifi system.
This unit will drive two 10" (wired in parallel thus 4 ohm) for channel, the supply voltage is +/- 56V with a 600VA toroid transformer, the estimated output power will be around at 200 Watt RMS @ 4 ohm.
The max power to keep the excursion of the woofers within XMax supported from each 2x10" is around at 120 Watt RMS (140W is the thermal limit) so I have much more power available from the amp than the power supported from the cabs.
But this is not a problem because I know that my ears will start bleeding much sooner than reaching the max power for my cabs !
So IMHO I wouldn't focus on the max power of your loudspeakers vs power of the amp, unless we are talking about pc speakers !
And I wouldn't worry about the exact real max power of the amp, I'm sure it is enough for a "safe" listening level with your cabs.
Regards,
Roberto
This unit will drive two 10" (wired in parallel thus 4 ohm) for channel, the supply voltage is +/- 56V with a 600VA toroid transformer, the estimated output power will be around at 200 Watt RMS @ 4 ohm.
The max power to keep the excursion of the woofers within XMax supported from each 2x10" is around at 120 Watt RMS (140W is the thermal limit) so I have much more power available from the amp than the power supported from the cabs.
But this is not a problem because I know that my ears will start bleeding much sooner than reaching the max power for my cabs !
So IMHO I wouldn't focus on the max power of your loudspeakers vs power of the amp, unless we are talking about pc speakers !
And I wouldn't worry about the exact real max power of the amp, I'm sure it is enough for a "safe" listening level with your cabs.
Regards,
Roberto
Last edited:
This might help,
The Signal Transfer Company: Power Output
remember that the Lateral Mosfets have a current rating rating of only 7 amps in normal form and 14 amps in the "double die" types.
In practice they are very tough and not prone to failure in normal use.
You might be surprised just how little power you actually need. If you try this you will end up with a realistic figure needed for your amps,
http://www.diyaudio.com/forums/mult...much-voltage-power-do-your-speakers-need.html
The Signal Transfer Company: Power Output
remember that the Lateral Mosfets have a current rating rating of only 7 amps in normal form and 14 amps in the "double die" types.
In practice they are very tough and not prone to failure in normal use.
You might be surprised just how little power you actually need. If you try this you will end up with a realistic figure needed for your amps,
http://www.diyaudio.com/forums/mult...much-voltage-power-do-your-speakers-need.html
Very helpfull, thanks, I'll keep reading.
I don't really need that power if the amp where for indoors, an hifi for example.
This was a guitar amp/cabinet mainly to play in gigs, at home I play through a small amp, I think it's 15w or something like that, for practice and not to disturb neighbours this amp do the trick, but when rehersing with the band I have to compete with a real drumkit which gives 105dB of sound level pressure (i forgot the name in english for this unit).
The speakers are two celestions rated 60-70w 8 ohms with about 97 dB of sensibility, 3dB more efficient than usual guitar loudspeakers, and it's really loud, right now I have a chip-amp (I didn't know that until I had to clean the pots and saw no output stage but a chip which name I forgot), this amp gives around 100w on my cab, and It's ok, but the amp built several years ago with the hitachi mosfets gave the amp that "I don't know", I'm looking for that sound again.
Why am I obsessed with power output? Well, usually on gigs and rehersals you have to share the amp with more musicians, when musicians are involved (myself included) they/we like to push the amp, in gigs the amp maybe is not right in your face but maybe 4-6 meters away, so you can't hear yourself playing because you're too near the drums, the singer's monitor, the sound through your monitor is very low, whatever, you go as fast as posible and give a push to the volume knob... and not nicely, you're in the middle of a song... 6 meters away you can hear now the guitar but you really don't know if you're driving too much the speakers, well, in real gigs the guy in the mixing desk will warn you about this, but not always.
Beware of musicians!!
Sorry for the long reply and all my mistakes writing in english.
Kind regards.
that's where your "sound person" earns his keep.
He gets the best sound available from your equipment in that venue.
He is the person who will blow up your gear or save it from damage. He is the person that ensures your audience can "hear" what you are playing.
I'm a musician too (bass player) and "pro amp" units are made with a different "target" than amps for hifi.
A pro amp must support high volume for a long time (especially for bass amps where to compete with a 50 Watt guitar amp we need at least 200/250 W !!) so I think that using a P101 "low power version" is not safe for it's duration for the purpose that you want to do.
I think you need the high power version that doubles the output transistors so that you have good margin of safety (and the output transistors last longer).
Regards,
Roberto
A pro amp must support high volume for a long time (especially for bass amps where to compete with a 50 Watt guitar amp we need at least 200/250 W !!) so I think that using a P101 "low power version" is not safe for it's duration for the purpose that you want to do.
I think you need the high power version that doubles the output transistors so that you have good margin of safety (and the output transistors last longer).
Regards,
Roberto
Last edited:
The sound person will warn you in order to protect their mics, usually sm57's in front of one speaker, but he really doesn't know if you can hear yourself unless on stage there is another guy on the monitors mixing desk....
.... It's Ok then... as long as he is on stage (not in the bar) and can understand your "body language".
I've seen gigs where the mixing desks were on "stand-alone". He is the person that ensures the audience can hear what you're playing... or just the opposite, on stage you're listening yourself ok and a guy approaches the stage and goes "Are you really playing? I can't hear your guitar out there ". Then you want to kill the man... he is ordering a "cuba-libre" and the "on-stage-person" is nowhere to be found. Finally you give your amp a boost out of anger... your singer blames you because the guitar is now too loud, the guy on the audience gives you "thumbs up", now he can hear you...and then the mixing desk is manned again, the on-stage guy (usually off-stage) claims you're going to burn the mic, the only way to get their attention man, real story and that's why I don't want to burn my cabinet, as I said beware of musicians.. and really bad "sound engineers". Off topic, but I hope you get the point. In an anger attack I don't want to blow up the speakers, and don't want anyone else to do it for me.
.... It's Ok then... as long as he is on stage (not in the bar) and can understand your "body language".
I've seen gigs where the mixing desks were on "stand-alone". He is the person that ensures the audience can hear what you're playing... or just the opposite, on stage you're listening yourself ok and a guy approaches the stage and goes "Are you really playing? I can't hear your guitar out there ". Then you want to kill the man... he is ordering a "cuba-libre" and the "on-stage-person" is nowhere to be found. Finally you give your amp a boost out of anger... your singer blames you because the guitar is now too loud, the guy on the audience gives you "thumbs up", now he can hear you...and then the mixing desk is manned again, the on-stage guy (usually off-stage) claims you're going to burn the mic, the only way to get their attention man, real story and that's why I don't want to burn my cabinet, as I said beware of musicians.. and really bad "sound engineers". Off topic, but I hope you get the point. In an anger attack I don't want to blow up the speakers, and don't want anyone else to do it for me.
Hi Roberto, I completely agree with you, I had this fight with the bass player on my band, he bought a 100w amp, I said that he will need around 300w to get on top of guitars. No way, he said. So one day we decide to give a benefical and go only with amps and a pro amp for the vocals.. then he realized what i was telling him for about two months.
I know the p101 is maybe not enough, but my old one wich used the same pair worked for months, I explained this before.
I will test it in order to buy another PCB and build a stereo amp , 4x12" but stereo wired... another headache for the "sound man"... man, put two mics... I don't have an extra mic... then get the stereo from the DI box... -What is a DI box? real stories. Another good one (I was working as the sound man this time), I was checking the gear.. oops.. no noise gate for the mics??... let alone compressors or anything fancy.... -Hey, how come you don't have noise gates for all those mics on stage? -Well, we really don't make too much noise... 5 x 200w monitors on stage.. not too much noise he said.
Thanks and sorry for all the pointless stories. I hope some of you get a smile from these stories.
New idea.
Maybe I could approach this in a different way and without the pointless anecdotes.
I have 2 speakers rated 60-70wrms, 8 ohms each, my idea is now to build an stereo.
What if I rewire (and modify) the cabinet for stereo and give a p101 (new transformers, 30-0-30 160VA each) as Rod recommended to me for each speaker?
Old power transformer will go to the box and I will need one of 320VA+ or two of 160VA.
Have you found one way to be better (I mean more efficient, less prone to failure, simpler design) than the other?
I also read that toroidals from 300VA needs an specific circuit, some kind of low-turning-on.
Thanks and happy weekend to all of you.
Maybe I could approach this in a different way and without the pointless anecdotes.
I have 2 speakers rated 60-70wrms, 8 ohms each, my idea is now to build an stereo.
What if I rewire (and modify) the cabinet for stereo and give a p101 (new transformers, 30-0-30 160VA each) as Rod recommended to me for each speaker?
Old power transformer will go to the box and I will need one of 320VA+ or two of 160VA.
Have you found one way to be better (I mean more efficient, less prone to failure, simpler design) than the other?
I also read that toroidals from 300VA needs an specific circuit, some kind of low-turning-on.
Thanks and happy weekend to all of you.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.