JLH 10 Watt class A amplifier

AndrewT said:
A 400mW 1k0 trimmer has a maximum current rating of 20mA.
I would never pass more than 14mA through and probably try to ensure <=6mA (9%) for normal operational use and 10mA (25%) for transients.
If your math is right, then your wording is confusing. It sounds to me like you're saying 25% of 20mA is 10 mA.
But seriously though folks, Andrew is right about how to decide the power rating for a pot: convert to current = sqrt(P/R) and the current rating is good for the full range of the pot.
 
the rating for the resistor is a power rating.
That applies to both fixed resistors and variable resistors.
If a 600mW rersistor runs at half it's rating. I think most folk would understand that to mean half power, i.e. 300mW.

If that same resistor is run at half current then one is using 25% of it's power rating.

It's the same for a variable resistor.

Running the track at half current is 25% of power rating.
6mA is {6/20}^2 =9% of rating. {10/20}^2 = 25% of rating.

It is not usually necessary to spell out power when referring to rating.
 
Thanks for clarifying that. I don't think it's all that obvious. You used the word 'rating' for both power and current.
A 400mW 1k0 trimmer has a maximum current rating of 20mA
Another suggestion: rather than just giving out numbers, you would assist many readers greatly by showing the formula(e) you used to derive them. In this case, don't assume the reader knows how to derive current from power and resistance. Your posts (from what I've seen) are knowledgeable and accurate and others can gain from your assistance.
 
Re: higher voltage

kimarin said:
Yes I was thinking of using this version ( http://www.btinternet.com/~tcaas/jlhupdate.htm) but after a tip from where I bought my amplifier case, I'm going to try the 2SC3264. Only 2 pr. channel instead of 4. I wonder if I can use a 25v-0v-25v 500va trafo instead of a 18v-0v-18v 500va trafo to raise the output a bit?

Better still try Sanken 2SC2837 outputs or even better 2SC2526 Fujitsu ring emitter transistors.
 
Re: higher voltage

kimarin said:
... Only 2 pr. channel instead of 4. I wonder if I can use a 25v-0v-25v 500va trafo instead of a 18v-0v-18v 500va trafo to raise the output a bit?
It will depend mostly on the size of your heatsinks and where you set the bias current. Don't underestimate how much heat a Class-A amp gives off. My basement room is noticeably warmer when my amp is on.
 
Fanuc said:
Key Change: Replace 2SC3421 or 2N1711 with 2SC4736

an 2SC4736 as it has almost 10x the gain of the the other two. Gain Linearity is ruler flat also. Ft/Cob is not too high (but relatively low) fortunately as you would get oscillations with video transistor types.

Should give enormous improvements to the output stage, more than anything I've ever seen since the '96 design was published.

Kevin
 
Just letting you know I will be grasping the power supply of this amp quite amplely. A bootstrapped AD797 based jung regulator with neutralisation capacitor (a meticulous copy of Jan's PCB layout except with the T03P output transistor.) and some obvious omissions. 🙂

Full S.M. would solve the problems but who can use it?

Take Care and god bless.

Have me Kef Reference Model 4's coming tomorrow. Be interested how they compare with electrostats.

Kevin
 
Fanuc said:
Just letting you know I will be grasping the power supply of this amp quite amplely. A bootstrapped AD797 based jung regulator with neutralisation capacitor (a meticulous copy of Jan's PCB layout except with the T03P output transistor.) and some obvious omissions. 🙂

Full S.M. would solve the problems but who can use it?

Take Care and god bless.

Have me Kef Reference Model 4's coming tomorrow. Be interested how they compare with electrostats.

Kevin
I understood very little of what you wrote.
 
paulb said:

I understood very little of what you wrote.

Yup, I will publish schematics in a few days. Let me get it working first. Have to include DC protection facilities also. Due to the regulator design and criticalities in it I will put the reg and amp on the same PCB. (essential)

Protection of excessive offsets (my plan is to use a servo anyway - no dc blocking cap) and full dc overload. Plus provision for P channel jfets. Not happy with the Toshiba variety (too much capacitance)
 
Fanuc said:


an 2SC4736 as it has almost 10x the gain of the the other two. Gain Linearity is ruler flat also. Ft/Cob is not too high (but relatively low) fortunately as you would get oscillations with video transistor types.

Should give enormous improvements to the output stage, more than anything I've ever seen since the '96 design was published.

Kevin

I suspect the Zin on the transistor (due to it's enormous gain) could permit bootstrapping the input Q. Improving Negative rail PSRR but getting the most out of low gain jfets.

The mosfet version could do this.
 
I emailed JLH's publishers (BH/Newnes) pointing out this fact.
I also pointed out the fact that it's been five years since JLH's passing.
I suggested a publication of some of JLH's past work. They have managed to do this for Self and Hickman, so why not JLH?

I didn't even get a reply. Tossers.

I am daily touched by the work of this Great Man.
I have all of his books and most of his articles.

RIP, JLH, legend and genius.
 
  • Thank You
Reactions: The Prof
Greetings JLH69.
Yes I had a similar experience trying to chase up a more complete bibliography of his material for Geoff Moss's site. The obits indicated that he did some RF articles for a couple of Ham or Shortwave mags and they just didn't want to know about any sort of enquiry into his work. At least I got a reply saying that they weren't interested! Sad.
I noticed you said you had a lot of his articles. I've got all the WW (EW) and HF-NRR material and Geoff Moss has kindly sent me some of the ETI work. Is it possible for you to contact me (thru' diyaudio members email) as I would like to chase up a couple of ETI articles and get better photocopies of a couple of PCB layouts that have not reproduced themselves clearly enough to work from. (The local Assie ETI, now defunct, was very different to the UK product so public libraries here are no use.)
The other thing I have noticed in the last few days is the limited recognition he has got in the US. Nelson Pass and a few others are cognizant of his ideas on feedback and appreciate him. But I am ploughing through a large work on audio electronic mysteries that seems quite oblivious to JLH's contribution to the discussion of amp stabilty/feedback/sound quality.
Greetings to Southern UK. We left Bristol in 1957. Might get back there soon with retirement looming.

Yours, Jonathan Bright
 
Work to be done

Well guys,

for the next couple of weeks I'll do some repair work on my two JLH 2003 amps.

As one has an oscillation problem it will be cured (in both amps) and speaker protection and power on/off thumbs suppression will be added.

A change in case design will take place as well.

Their width will increase from 24cm to 39 cm and front and back panel will go from clear to smoky.

It still makes me feel a bit 'lousy' to dissect this work. But, hopefully, I can put them back to the place they belong. Driving my Trio's..

Jos
 
Stormrider said:
How much heatsinking does Q3/Q8 need with +/-20-22vdc rails and 1A to 1.5A bias? Are those TO220 heatsinks Paul B. used necessary? I will probably use MJE3X0 or MJE1503X devices.

After my first testruns with only a "normal"to220 heaksink I found they got too hot. Blistering to be precise.

So I added a smaller second one at the front". That took temperature to touchable levels.

This is how it looks.

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


Jos
 
Re: higher voltage

kimarin said:
Yes I was thinking of using this version ( http://www.btinternet.com/~tcaas/jlhupdate.htm) but after a tip from where I bought my amplifier case, I'm going to try the 2SC3264. Only 2 pr. channel instead of 4. I wonder if I can use a 25v-0v-25v 500va trafo instead of a 18v-0v-18v 500va trafo to raise the output a bit?

Actually on reflection those 2SC3264's do look quite good. 200 watt types. Though I think the MJ3281's look better from gain linearity point of view.

I keep forgetting that 90% of T-O3P plastic audio types are useless in Class A amps when you put continous current through them. Thermal resistance of Junction to case etc. There designed for class B amps. The MT-200 case of Sanken looks an exception for plastic types. Will look at some more of there devices.

PS. If you mean't in your post earlier, you can replace four 100w types with 2 of the 2sc3264 i doubt that will work (depending on current) probably only MJ3281 in T-03 package could do that and get very hot.

Jos (dutch diy)

Those heatsinks look undersized compared with the ones fitted on the PCB's I got from EW when the JLH article appeared. In particular the CCS one. (PNP)

Best Regards

Kevin
 
Hi All
been some time since i posted here i have been going the build this amp for a long time now but this has been mainly due to my lack of know on setup testing basically. Apart from the fact i have a single ended valve/tube 2a3 Amp now but still interested in this Amp but would like to have the PCBs for it if anyone is selling them on here also what maybe of interest to some of you

Have the schematic for the headphone version that Geoff kindly send me which is just a scaled down version of this JLH amp this is not the headphone amp on his site if anyone is interested in a headphone version i may start a new thread on this if i get permission to post the schematic from Geoff a mini JLH amp for you cans sounds good to me and cheaper too
😎 😎 Colin