Slewmaster - CFA vs. VFA "Rumble"

OS
Thanks. Basicly the thermal performance of the OPS with drivers at separate heatsink should be better than output+drivers at main heatsink ?

Sorry that I have to ask such questions but I do not have much experience with EF3.


I only have one experience with EF3 (3 pairs), I have used single transistor V bias spreader MJE340 and amp went up to 170C deg before I turned it off (still alive heh).
Than I swapped MJE340 with BD139-16 and amp was very little undercompensated.

THX
 
OS
Thanks. Basicly the thermal performance of the OPS with drivers at separate heatsink should be better than output+drivers at main heatsink ?

Sorry that I have to ask such questions but I do not have much experience with EF3.


I only have one experience with EF3 (3 pairs), I have used single transistor V bias spreader MJE340 and amp went up to 170C deg before I turned it off (still alive heh).
Than I swapped MJE340 with BD139-16 and amp was very little undercompensated.

THX

This one is a "hair" over comp. , my 15mv bias will drop to mid 14's when real hot.
I've run from 20-50+ C can not blow it up.
Ran the MT-200's at 500ma bias just to see what would happen. , hit 60C+
(could not touch it) .... still 14.5mv.
You are right ., this EF3 is fast to comp. Load heats the smaller driver Vbe first.

OS
 
Great learning stuff from you OS🙂 both regarding grounding and regarding outputstage🙂
So you choose to have fast tracking over mounting all the output and drivers on same heatsink?

Can you also make a PDF document with grounding for slewmaster and Kypton C?
( only for private use I'm asking)
 
Great learning stuff from you OS🙂 both regarding grounding and regarding outputstage🙂
So you choose to have fast tracking over mounting all the output and drivers on same heatsink?

Can you also make a PDF document with grounding for slewmaster and Kypton C?
( only for private use I'm asking)

I suppose Harmon Kardon actually developed this Vbe "trick". Theirs had NO
heatsink for the drivers and the Vbe "sandwiched" between the those bare drivers.
This was before the next generation of receivers (HK990) that actually do
use a pair of ON semi Thermaltrak's. (instant tracking).

The Slew OPS is a compromise between the fast tracking and the higher currents
required by a much larger amplifier.

I did solve the last -110db hum issue. The AC line fields were getting picked up
by both the Kypton and spooky. (extremely low level)
I am making shielded cables out of old parallel (round) IDE HDD cables (with a solid copper ground added.).
Trafo primary leads and the main AC mains input get these.
I'll get my -120db ... I don't give up !!!😀

OS
 
Thanks OS for explenation.
The last question, if I will use bd139 instead of mje340 in your's ops would that be ok or leave as it is --> I have bd139 nice isolated versions and is easy to bolt them on 😀 (I am bit lazy)

?? I have not tried the BDxxx . KSC3503 = much over comp. and
the MJE340 = nearly perfect (very slight overcomp.)

On the slewmaster OPS ....
You CAN trim the OP co-efficient with R103 (1.8K) in the Vbe circuit.
R104 (1.2K) can fine tune the drivers co-efficient , as well
OS
 
Grounding "101" continued.

Absolute bliss.

If you use a 5U chassis as big as a breadbox 😀 for your amp , these tips
might seem "anal"...

But ..
#1 - AC line fields.

With distance (and the "breadbox" enclosure) , all might be well.
If you have even a tightly twisted line pair , 4-5CM will still get you evil hum. A simple and cheap way
around this is to use old PC hard drive (round shielded) cables. (below 1)

- make a litz braid with the 2 AC and a 20ga sanded magnet wire. Snake that
through the PC cable shield , and you will have a cable that can lie on top
of an input stage while inducing NO hum !! 😱

- Your trafo primaries and secondaries can also benefit from this (below 2).
You can tie all the grounds together to happily direct connect to your chassis
ground (main earth).

#2 -Toroidal fields.
Even in the "breadbox" enclosure , the section of the otherwise perfect toroid
that has the primary -.secondary takeoffs (below 3) is where "the force" IS disturbed.
Aim this EMF "GUN" at a sensitive amplified point .... and you will hear it !
This is new to me !!... was told this was impossible - NOT !!


All this info is readily available , used by precision instrumentation manufacturers
and is often ignored by DIY'ers.
Do it , and you will have the purest , quietest amps out there !!🙂 (-115 / input
semi noise ONLY).

OS
 

Attachments

  • grounded AC.jpg
    grounded AC.jpg
    145.4 KB · Views: 676
  • grounded trafo.jpg
    grounded trafo.jpg
    166.7 KB · Views: 645
  • The force.jpg
    The force.jpg
    139.4 KB · Views: 568
Last edited:
OS

I have another small question regardless HK's bias spreader. I do not know how in real life but in simulation the OPS bias is strongly dependent on VAS bias stability.
Can I just add R22 as on the drawing bellow ?
In simulation it helps a lot - VAS bias drift does not affect much OPS bias.

THANKS
 

Attachments

  • bias spreader.png
    bias spreader.png
    42.5 KB · Views: 540
Hi all! This thread is getting very long and increasingly difficult to glean the wheat from the chaff. To a beginning builder of the slewmaster is there a post that summarizes recommended heat sink area and where are the thermal compensation elements that have to integrated with the output devices. Leach used diodes. Should all the p and n output devices be on one heat sink or can they be divided on two heat sinks (pnps on one and npns on the other). I see the p-s on one side and the n-s on the other side of most pc layouts. Are these heatsinks on ebay big enough for a + and - 50v rail amp. 3 Aluminum Heatsinks 2" H x 1 63" w x 6 75" Gold Chromate Plated 65S98 14 | eBay Thanks, Ray
 
OS

I have another small question regardless HK's bias spreader. I do not know how in real life but in simulation the OPS bias is strongly dependent on VAS bias stability.
Can I just add R22 as on the drawing bellow ?
In simulation it helps a lot - VAS bias drift does not affect much OPS bias.

THANKS

Yes , Self magic resistor.

As far as this VAS bias drift , my IPS's drift very little ... .1/10th to 1/5 mA.

OS
 
Hi all! This thread is getting very long and increasingly difficult to glean the wheat from the chaff. To a beginning builder of the slewmaster is there a post that summarizes recommended heat sink area and where are the thermal compensation elements that have to integrated with the output devices. Leach used diodes. Should all the p and n output devices be on one heat sink or can they be divided on two heat sinks (pnps on one and npns on the other). I see the p-s on one side and the n-s on the other side of most pc layouts. Are these heatsinks on ebay big enough for a + and - 50v rail amp. 3 Aluminum Heatsinks 2" H x 1 63" w x 6 75" Gold Chromate Plated 65S98 14 | eBay Thanks, Ray

Way too small. OPS needs a 3U (5"+) height 2" X at least 11" long ,
for 90 degree IPS mounting (like mine). .5 C/W ...or mine is about .4
(larger than the HS USA 12").
.5C/W will just run warm with heavy use.
PS - i'm using one of those 170$ chinese Ebay chassis's (3U).

OS
 
Last edited:
Almost 1/2 million Views ...

Would be nice if DIYA gave me a "21'st century first post" !! 😉

- I have all the necessary layout/grounding/thermal info.
- I have the symasui , wolverine , spooky , and kypton C "tweaked".
- all the best choices for Vbe / output (thermal) totally rock solid.
-I've even rounded up all the gerbers and other info partaining to the slew series.
-All the master simulations are updated and tweaked. (layouts , too) - all
in one neat space of my HDD !!
..... (LOTS of material - 100m/b+).

The amps are quite flawless and can hit -115+ noise/hum 20-30ppm
full power and will "thump" any speaker hooked to them. (VERY durable).

My problem is to update and have a central depository for all this info (Like
GR hughes pointed out ).
- A PDF for grounding , with all the finer tricks.
-General construction tips for the OPS's / power supply.
- PDF for each IPS.
Have Camera , have the real amps/IPS .... but no ability , other than to
link with text to the post containing many attachments and updates.

PS -Below #1 ... shielded rail wires drop hum below slightly modulating the -115db
noise signal (below -115). Lay the rail cable right on top of an IPS - NOTHING.
#2 - amp playing Alan parsons the best I've ever heard. ( A 5K$ amp) 😱

OS
 

Attachments

  • grounded rails.jpg
    grounded rails.jpg
    161.2 KB · Views: 295
  • playing-happily.jpg
    playing-happily.jpg
    172.7 KB · Views: 321
why don't you build a web site for this ?
I can help-you on this as i'm too a web programmer.

You see the fidelityforce under my signature , I forgot that Daniel does hosting 😎
Any help would be appreciated. I'm rusty on PDF creation.

But , I do have all the hard stuff done (designs , PCB , the " tweaks").

I'll have to PM Dan about that.
Edit - I saw what he does have - just direct links to the posts (but he has what he has right ?)
I don't know why DIYA won't , builders buy store cases and I help badger
diy'ers ?? (mo $$ for jason) ?

OS
 
Whatever you are familiar with, in order to put things in shape, (Word or pdf etc...) i can translate-it in HTML for you.
Just the matter to not lose too much time discussing look and design.
And even, if you trust my honesty* enough, upload them in any web site where you have an FTP, create the links in the menus etc...
Just PM me if you need, you will be welcomed.