JLH Power Survey

[astouffer]

To anyone running a JLH or similar amplifier. What voltage and current do you have the amp biased for? I plan on building a version with MJ15003 output devices and want to use a +/- 24v supply because of the availability of switching power supplies. One site has 24v 2amp supplies for $4.95. Four will cost less than half of what a similar toroid would run. Any HF noise should easily be filtered with some caps and maybe a choke.


Adam

 

[tlf9999]

I got a few of them but mostly running them at about 750ma off 36v rails.

Mine are mostly mosfet versions so they run most like class AB than class A.

 

[dutch diy]

J-amp

Well,

my J-amp runs on 29.5 V @ 2.1 A bias for the last 2 years.

It has 4 mj15003 devices per channel.

It was designed to run on a 3 A bias but I had to limit that because of heat build up in my audio cabinet.

Sound: I've had no real comparison made between both setting.

 

[dutch diy]

I plan on building a version with MJ15003 output devices and want to use a +/- 24v supply because of the availability of switching power supplies. One site has 24v 2amp supplies for $4.95.
Adam

Be aware that the current drawn from the powersupply is not a steady Ibias. It will run upto 2 times the bias current. Eg: No signal : only Ibias flows through the ouput transistor. On the 'negative' side of the sinus current will be just above 0 A ( [Class A ]. On the 'positive side of the sinus current needs to be double the Ibias.

 

[astouffer]

Thanks for the info. It turns out the place selling power supplies is sold out, guess I'll be building my own. Do you use any regulators to reduce hum or should capacitors and chokes be enough?


Adam

 

[Ipanema]

Hi guys,

I built JLH with this circuit http://www.tcaas.btinternet.co.uk/jlhupdate.htm.

The bias is set at 300mA due to small heatsink. I am very dissatisfied with the sound as compared to GC. Also substitue TIP3055 with MJL21194 with no improvement in sound. It has poor control on the lower end, rough top end and not as detail as GC. It only excel with solo female vocal. Can you tell me what's went wrong?

Thanks.

 

[Ipanema]

Oh yes, my DC supply is +/-26V.

Thanks.
😉

 

[Greg Erskine]

Hi Ipanema,

I haven't made a JLH yet.

On the schematic, it looks like you missed C6/C6A and C7/C7A. Rail filtering is important on amps with poor PSRR.

Feedback cap C4 may be necessary. I know it is missing off the "final" schematic.

I don't believe running the amp at low bias is a good test for a class A amp. Put some big heatsinks on, and/or use a fan.

You need to post the details of your power supply. It needs to be reasonably ripple free.

A picture of the amp may be useful too.

Regards

 

[tschrama]

300mA is too low for such a low feedback BJT design... I run mine from 30Volt single rails at 1.4A bias.. I also build a MOSFET version that sonicall out performed the BJT, regardless of bias...

 

[Ipanema]

Hi Greg,

The local supply decoupling caps are 1000uF + 100nF (ceramic). Power supply circuit is capacitance multiplier as suggested here :

http://www.tcaas.btinternet.co.uk/jlhcapmult.htm


I use Figure 2 which suppose to sound better without C4 albeit higher offset. Offset are 10mV and 80mV for my stereo pair.

Will try higher biasing when I got a macho heasink. 😀

Do you think that BC560C is inferior than 2SA970 in this application?


Thanks.

 

[Greg Erskine]

Well Ipanema, it sounds like you are on the right track. It may be just the bias setting then.

I've never compared BC560C to 2SA970 so I'm not in a position to comment.

 

[tlf9999]

The bias is set at 300mA due to small heatsink.

the bias is too low. depending on how loud it is, you ought to run it a little bit higher. I would suggest at least 750ma but hopefully over 1amp.

You may run it at lower voltage (like just off the +24v to reduce dissipation). alternatively, you can use mosfets as output and run this in class AB.

 

[MikeW]

I am using +/-17 volt rails with 1.7 amp current draw. Dual MJW 21194 for the outputs.

 

[dutch diy]

J-amp power supply

Do you use any regulators to reduce hum or should capacitors and chokes be enough?
Adam

Yes, I do:

An externally hosted image should be here but it was not working when we last tested it.

 

[Ipanema]

Thanks. Will try higher bias. BTW, what you guys think of JLH sound as compared to GC? and also the sound between variation of JLH e.g. 1969, 1996 and 2001.

😀

thanks.

 

[tlf9999]

JLH vs. GC: the GC is considerably louder. and at low volume, there isn't much difference between the two. When pushed over its limit, the GC starts to sound harshier more and sooner than the JLH. When overdriven, the JLH tends to sound muddy rather than harsh.

1969 vs. others. I never tried other versions as laid out. My interest is mostly on the 1969. and I have tried the stock bootstrap, without the bootstrap and with a CCS load. Again, not much difference when working in their parameters. However, the one with a CCS tends to sound harshier when pushed. and no-bootstrap version is quite gentle when pushed.

For what I do, I decided to stay with the no-boostrap version.