John,
I assume by MJL3821A you mean MJL3281A. I'm tempted to try this. Which circuit are you using and have you compared the MJL3281As with MJ21194s in that circuit?
For the record the hum I initially experienced using MJ21194s disappeared when I rebuilt the circuit using them in a better layout. They sound marginally better than the MJ15003s to my ears.
Tim
I assume by MJL3821A you mean MJL3281A. I'm tempted to try this. Which circuit are you using and have you compared the MJL3281As with MJ21194s in that circuit?
For the record the hum I initially experienced using MJ21194s disappeared when I rebuilt the circuit using them in a better layout. They sound marginally better than the MJ15003s to my ears.
Tim
Hi Tim
Yes should have been MJL3281A's. I used them in the original circuit with a bootstrap load resistor, but simulations using the directly coupled output mod with current sources suggests that the same compensation should work in this version too. They make a bigger difference in the bootstrap circuit in simulation at least because the rise time is limited by the frequency response of the upper output transistor (bootstrapping itself). With a current drive it should be faster, but these transistors are pretty fast anyway.
From my 100W Class AB design I'd agree that MJ21193/4 pair are better than MJ15003/4 (slightly faster again). But I think the MJL3281A's should be better still.
John
Yes should have been MJL3281A's. I used them in the original circuit with a bootstrap load resistor, but simulations using the directly coupled output mod with current sources suggests that the same compensation should work in this version too. They make a bigger difference in the bootstrap circuit in simulation at least because the rise time is limited by the frequency response of the upper output transistor (bootstrapping itself). With a current drive it should be faster, but these transistors are pretty fast anyway.
From my 100W Class AB design I'd agree that MJ21193/4 pair are better than MJ15003/4 (slightly faster again). But I think the MJL3281A's should be better still.
John
Hello all,
Anyone have any advice on running the High Power JLH (two pairs output devices) with 34V rails? I have a home built Aleph 5 that I not using, but would like to pull the A5 and drop in the JLH module. I have to stick with SE inputs on the JLH (Moss PCB)
I was impressed with the JLH with 24v rails...
Anyone have any advice on running the High Power JLH (two pairs output devices) with 34V rails? I have a home built Aleph 5 that I not using, but would like to pull the A5 and drop in the JLH module. I have to stick with SE inputs on the JLH (Moss PCB)
I was impressed with the JLH with 24v rails...
What a lovely looking amplifier you have made ? the Rhino or JLH ? in your signature.
Sounds Good, The first watt F5 Turbo To me sounds better than the original F5, the Voltage was increased i think and also two pairs of output devices added, Good luck!
Im wanting to run this amp from a 12v source basically car battery with another more efficient amp for the bass. If needed the voltage would be stepped up but need to look into the pros and cons of doing so.
Someone before has said they had the JLH working on 12volt fine without upping the voltage.
Im wanting to run this amp from a 12v source basically car battery with another more efficient amp for the bass. If needed the voltage would be stepped up but need to look into the pros and cons of doing so.
Someone before has said they had the JLH working on 12volt fine without upping the voltage.
Unless you use two amps in balanced working mode you will only get around 1.5 watts into 8 ohms from a 12V supply - this may just be ok for mids & treble but that depends upon your speakers and your preferences.
Pretty low Realy inst it! yeah I was told that Class-D amps would sound good in Car audio and many many times more efficient.
Hmm yeah balanced would be fairly easy to achieve, I was hoping to get around 5 watts.. oh dear lol
It is a JLH amplifier, but i already have a few smaller versions of it named as "JLH" on the front panel so "Rhino" is just for not repeating the same decal on all the amplifiers
And because it is big So it is a pure JLH1969 amp here.A smaller version (for me it gets too hot, but working
) Project | Homebuilt Hi-Fi - A user submitted image showcase of high quality home built hi-fi components.
Good evening dear colleagues,
I opened a thread with these questions but didnt get so much help so i hope that someone that builded this wonderfull amp could help here.
I have bought these two radiators from the picture attached and hoping to find some answers from you guys:
*Note that the dimensions for both of them are: L=25.5 cm; l=23.5cm; h (for the base)=1 cm. The dimensions for the fins are: L= 23.5cm and h=2.5 cm. The black radiator has one fin less than the other.
1. Are these results ok for the total dissipation area?
Abase=2(L*h+l*h+L*l) =>Abase is 648.25 square cm + 885 square cm for the total of 15 fins (from the formula Atot=L*l*Number of fins) => A=23.5*2.5*15?
The total dissipation area is, from the calculation above (for the black radiator) 1533 square cm?
2. Are these radiators suited for a stereo JLH for QUAD ESL at 42W@8Ohm , @+-28V cc @3.2A quiescent current with 4 pieces 2n3055 per module?
3. Wich solution is best for the silver radiator: to spray it with a black thin coat of thermo resistant paint or to black anodise it (sollution wich i do not have at disposal in my county)?
Thanks in advance
I opened a thread with these questions but didnt get so much help so i hope that someone that builded this wonderfull amp could help here.
I have bought these two radiators from the picture attached and hoping to find some answers from you guys:
*Note that the dimensions for both of them are: L=25.5 cm; l=23.5cm; h (for the base)=1 cm. The dimensions for the fins are: L= 23.5cm and h=2.5 cm. The black radiator has one fin less than the other.
1. Are these results ok for the total dissipation area?
Abase=2(L*h+l*h+L*l) =>Abase is 648.25 square cm + 885 square cm for the total of 15 fins (from the formula Atot=L*l*Number of fins) => A=23.5*2.5*15?
The total dissipation area is, from the calculation above (for the black radiator) 1533 square cm?
2. Are these radiators suited for a stereo JLH for QUAD ESL at 42W@8Ohm , @+-28V cc @3.2A quiescent current with 4 pieces 2n3055 per module?
3. Wich solution is best for the silver radiator: to spray it with a black thin coat of thermo resistant paint or to black anodise it (sollution wich i do not have at disposal in my county)?
Thanks in advance
Attachments
Should just be enough. It'll get hot. The best way to work this out is to measure it - get a 10W or 20W metal cased resistor, bolt that to your heatsink (radiator); do some measurements and run the math.
On colour - the best thing to do is to leave the silver radiator (heatsink) as it is. Try and work out a physical arrangement which allows for the colour difference
Edit - this sort of resistor
An externally hosted image should be here but it was not working when we last tested it.
. Or this one 
An externally hosted image should be here but it was not working when we last tested it.
(VISHAY SFERNICE - RTO020F12R00JTE3 - RESISTOR, 20W )
Thank you all for the sugestions. Now i am thinking about a measuring method: atach the radiator to a smoothing iron with an ampermeter conected in series, and then see how much wattage is applied to the sink.. What do you think about this method?
What do you think its best, to use these rads for the same amp but with 3.7A@+-22 V cc@ 28W (4 tranies per amp for better bass or use the variant with 2 tranies per amp? Wich variant do you recomend? The one with paralelled outputs has better control over the speaker, but i dont know how will sound the variant with a single pair of outputs.
Thanks in advance.
It'll be a rough approximation - the smoothing (clothes?) iron will also lose heat in other directions.
Another method would be to heat the radiators up to a known temperature (say 100C in the oven); move to an outside space (24C) and time how long it takes to drop to perhaps 60C. And do the math- you can weigh the radiator and you know it's aluminium so thermal mass can be found.
the more data points the better as you're trying to measure an integration curve.
I am a little "old school" - the moment you add extra output pairs you're into device-matching purgatory. If you can live with less watts, the simpler architecture is always* going to be better. And easier.
* I was trying to think of a counter example but haven't yet. Anyone? Ferris?
Thank you very much for the answers. I was trying to aproximate the temps with the smoothing clothes iron..
I guess i will stick to one pair of outputs as you said, because novadays we cannot find motorolla 2n3055 to closely match them. In Romania i have found some fake ones and another batch made by IPRS Romania at the local store...
