ASR Emitter II (1990) repair and restoration

[Rytikar]

The Introduction.

Hi all,

I've been involved in HiFi since I was 14 buying my first receiver/tuner/turntable set. I later moved on to a Pioneer SA-9800 integrated amp with TX-606 tuner and a Technics turntable feeding Pioneer HPM-110 speakers. I wished I had kept all of those (especially the HPM-110), but you know how it goes.

Being a young lad with no responsibilities and too much money on my hand I bought the ASR this thread is about. The year was 1990, and I got it together with an Ariston turntable and an ASR Microbase phono preamp. (More on the Microbase later, perhaps).

I listened to this setup for quite some time (10-12 years?), changing to Monitor Audio PMC703 floorstanders (sand filled, of course).

Well, life happened, I moved to another country and the HiFi interest moved into the background. Got rid of my turntable (damaged) but kept the Microbase and the records. Used AV setups for some time. The ASR was too heavy to lug around (more on that later), so I gave it to a good friend that at that time had problems with his own setup. He used it for a good 15 years. We stayed in contact, visiting once in a while.
Last time he told me that the Emitter had died, at least one channel. He bought some nice NAD stuff and placed the ASR into the corner where it waited silently, gathering dust.

Meanwhile life changed again with more time and more space available. I realized that something was missing and revived my long forgotten electronics hobby ending up repairing amplifiers and other HiFi gear. With that the interest in music and the technical aspects rekindled. I've repaired dozens of amps in the meantime, swapping gear whenever something nice comes in and gets fixed. Liked NAD 3020 a lot, used a Thule IA100 a while (nice one) and actually built myself a KT88 single ended tube amp lately. With that I switched from my little Sonus Faber Toy speakers to some Klipsch, we're talking about 8 watts out of the KT88 here.

The time was in to tackle the big one.

So, here I am, having just revisited my old friend, this time telling him in advance to make the Emitter travel ready. I lugged it home and without even turning it on began to investigate.
My friend demanded to be updated on the progress, so I wrote a long mail to him about the status.
Went to the holidays when the thought came to me that maybe the nice people of AK are interested in this repair, too.

If there is interest into the topic I will follow up with more info on the Emitter and where I am right now doing the repair, what I've done so far and what the plans are going ahead. I think the damage can be fixed (if not economically, but that's not the point, right) and the mighty Emitter be restored to its former glory.

Btw, the picture to this post is not my system but from the net. It shows an ASR Emitter II and its PSU from around the 1990s and is very close to mine. I took mine apart before taking pictures of the whole system. In the following posts I will have more of my own.

 

[eizner23]

250 watt what a beast! Blown output transistor would be a common fault to check for.

Here is the service manual, create an account and download it.

 

[Ds2000]

I had one a couple years ago. Emitter II Exclusive Blue. There were four parts, 303 pounds.
I’ll be watching this thread with interest.

 

[Rytikar]

The Patient.

ASR Audio is a german high-end manufacturer founded in 1980 and still exists (https://asraudio.de/).

They have built the Emitter I and the Emitter II since, adding phono preamps and power cords to their portfolio.

The Emitter I is the 'smaller' brother of the Emitter II. They share the same ingredients with the Emitter I having 'less' power supply and less output FETs.

The Emitter II has had its first release in 1980 and has been produced with several enhancements and major circuit changes ever since. From the outside they all look the same, though.
Two big heatsinks on each side connected with a steel frame and covered in acrylic plates to give a glimpse of the insides. There use darker or transparent acryl. Mine has the darker one, but I will cut a transparent top plate when the amp is finished.

The single PCB dividing the inside of the amp into an upper and lower part is built in a dual mono approach (and luckily so). Most of the capacitors are MKT. There are actually not that many electrolytics on the main board.
On the front you find the on/off switch with two power settings, a source selector and a tape monitor siwtch.
Of interest is the volume 'potentiometer'. It is actually a discrete resistor switch featuring matching high accuracy resistors. It clicks into the different settings and is quite fine grained. Looks like that got changed later.
Five high level inputs, one of them tape in/out and four massive speaker terminals are on the back.

Only the first version had an internal power supply, all following revisions come with external power supplies.
And it is the power supply ASR has it's focus on. Mine is a third revision from 1990. The PSU has four Philbert-Mantelschnitt transformers in the external cabinet, two for each channel. Four 25 amp metal cased rectifiers deliver the main rail. On the control board inside the PSU a small transformer is located as a standby supply. It's used to actually switch the unit on using a inrush limiting circuit on the board. Some minor rails are also generated on the PSU board, all delivered via a thick umbilical cord and industrial connector to the main amplifier.

The PSU does some pre filtering of the main rail using eight 4.700uF/100V electrolytics. You would think every normal amp would be happy to have ONE PM transformer with the pre filtering and do it's job just fine.
ASR thinks different, so inside the main amp, under the main board and normally out of view, are housed twelve 10,000uF/100V electrolytics doing the main filtering. (What this amount of capacity does to your output FETs when something shorts we will see soon in the next post).

This was only the beginning though. Recent iterations of the Emitter II have upped the game considerably. They come with two PSU cabinets with much more pre and main filtering capacity and in the final step adding a battery buffered supply in a third cabinet, so you end up with four units for this integrated amp.

The PSU weighs 50kg and the amp 47kg. It uses 10 FETs for each channel and delivers 250 watts into 8 ohm and 400 watts into 4 ohm (I've read 800 into 2, but not sure if mine is specified for that).

All this hardware gave some serious challenges to my workbench, my tools and my physic (asked my wife to get the PSU upstairs to 'the room'. She couldn't even lift it).

Next post we will look at the damage done and start the repair.

 

[Rytikar]

The Damage:

When I got the amp home I didn't even try to set power to it but started disassembling.

Very fast it was obvious what the problem was. All ten FETs of the right channel had burned up violently, some blowing their leads off.
All 100 ohm gate resistors for the FETs had got up in flames together with an assortment of neighboring bias and feedback resistors. The right channels protection zener diodes are shorted. The BY298 measures ok but will be changed. Don't know if the reason for the failure started in one zener shorting or one of the FETs and then cascading.

Driver- and bias transistors measure ok, but I will change them, too. They are available and cheap.

The board and the copper is thick and did not take damage. Some burned spots show on the upper PCB side after cleaning where burning resistors came to rest. It all cleaned up nicely considering the ordeal. Funny thing is my friend said he didn't notice anything just one channel not working anymore. That must have made quite some smoke.

Now the main problem, getting new FETs.

I've got into contact with ASR and Mr. Schaefer is very helpfull. I got service documents of what he recommends doing, some schematics, and help via mail.

The main FETs are Hitachi 2SK414 and 2SJ119. They are not in production since 1996, ASR don't have any stock left and their does not seem to be NOS available. I got me 10 of each from china to see what you get, but they look dodgy. Original SJ119 are green, 2SK414 black. All china FETs are black and Mr. Schaefer also thinks they are fake. He had tested lots of them, they will not work.

He recommends Toshiba SJ201 and SK1530 as used in later Emitters. Problem with those is they are quite expensive, you need to change both channels (20 FETs) and make considerable changes to the circuit to get them to work right. Bias, feedback and oscillation tendency needs to be addressed.

Now, the best middle way is using IRFP240/IRFP9240. ASR says they work very well, you just need to watch it while biasing as they react much faster than the old ones. The only circuit change to make is adding 0.22 ohm drain resistors to help with thermal runaway. Older Emitters need to change driver and bias transistors, too, but mine has the recommended ones already. So, that's what I will do, parts are ordered.

While I'm in there, I will recap (not the big reservoir capacitors, they measure fine). It's quite some work getting to things, you need to get the car tools out and the big soldering iron, too. Those big copper planes on the PCB suck heat like crazy. The big capacitors get out of the cabinet quite easy, they are mounted on acrylic plates and each can then be measured alone by taking the copper bars off. My LCR meter gulped a bit about 60,000uF combined. If this repair does not work out I at least can go copper trading.

I'm considering replacing all resistors and diodes on the other, working channel, too. Both for aesthetics and lifetime considerations. You can look into the amp after all. And it's best doing it now while the thing is apart.

I'm also thinking about replacing the LEDs. As mentioned I also have an ASR Microbase that has power LEDs like the Emitter shining red through the grates of the cabinet. They had been on for maybe 25 years usage straight. When I revived the Microbase lately they were very dim or completely dark.

Well, while waiting for parts I will look into the PSU, disassemble, clean and reassemble it.
More on that soon.

 

[Ds2000]

The Damage:

When I got the amp home I didn't even try to set power to it but started disassembling.

Very fast it was obvious what the problem was. All ten FETs of the right channel had burned up violently, some blowing their leads off.
All 100 ohm gate resistors for the FETs had got up in flames together with an assortment of neighboring bias and feedback resistors. The right channels protection zener diodes are shorted. The BY298 measures ok but will be changed. Don't know if the reason for the failure started in one zener shorting or one of the FETs and then cascading.

Driver- and bias transistors measure ok, but I will change them, too. They are available and cheap.

The board and the copper is thick and did not take damage. Some burned spots show on the upper PCB side after cleaning where burning resistors came to rest. It all cleaned up nicely considering the ordeal. Funny thing is my friend said he didn't notice anything just one channel not working anymore. That must have made quite some smoke.

Now the main problem, getting new FETs.

I've got into contact with ASR and Mr. Schaefer is very helpfull. I got service documents of what he recommends doing, some schematics, and help via mail.

The main FETs are Hitachi 2SK414 and 2SJ119. They are not in production since 1996, ASR don't have any stock left and their does not seem to be NOS available. I got me 10 of each from china to see what you get, but they look dodgy. Original SJ119 are green, 2SK414 black. All china FETs are black and Mr. Schaefer also thinks they are fake. He had tested lots of them, they will not work.

He recommends Toshiba SJ201 and SK1530 as used in later Emitters. Problem with those is they are quite expensive, you need to change both channels (20 FETs) and make considerable changes to the circuit to get them to work right. Bias, feedback and oscillation tendency needs to be addressed.

Now, the best middle way is using IRFP240/IRFP9240. ASR says they work very well, you just need to watch it while biasing as they react much faster than the old ones. The only circuit change to make is adding 0.22 ohm drain resistors to help with thermal runaway. Older Emitters need to change driver and bias transistors, too, but mine has the recommended ones already. So, that's what I will do, parts are ordered.

While I'm in there, I will recap (not the big reservoir capacitors, they measure fine). It's quite some work getting to things, you need to get the car tools out and the big soldering iron, too. Those big copper planes on the PCB suck heat like crazy. The big capacitors get out of the cabinet quite easy, they are mounted on acrylic plates and each can then be measured alone by taking the copper bars off. My LCR meter gulped a bit about 60,000uF combined. If this repair does not work out I at least can go copper trading.

I'm considering replacing all resistors and diodes on the other, working channel, too. Both for aesthetics and lifetime considerations. You can look into the amp after all. And it's best doing it now while the thing is apart.

I'm also thinking about replacing the LEDs. As mentioned I also have an ASR Microbase that has power LEDs like the Emitter shining red through the grates of the cabinet. They had been on for maybe 25 years usage straight. When I revived the Microbase lately they were very dim or completely dark.

Well, while waiting for parts I will look into the PSU, disassemble, clean and reassemble it.
More on that soon.

I’m not seeing any of the images.

 

[Rytikar]

Thanks for the tip, I checked with collegues and right you are, no pictures. I think I figured why. Will fix the pictures when I come home.

Ok, I hope I fixed it, I could only edit newer posts. But I will add the missing pictures from my second post here. Enjoy.

The volume 'potentiometer':

The opened, unfinished power supply:

On transformer as size comparison:

The underside of the amp with capacitors:

 

[Rytikar]

Fixed the pictures as good as I could. Should work from now on. Sorry.

 

[Rytikar]

The Power Supply:

At the end of the PSU restauration I got this comment from Mr. Schaefer when a problem after reassembly was solved.

Translated from german: "Oh, you're disassembling the power supply? Had I known this beforehand I would have strongly discouraged you. This is major work!"

And right he is, took me ten hours.

Now, there was nothing wrong with the PSU as far as I knew, so why take it apart? Several reasons, actually. Curiosity, cleaning, replacing the standby power caps, Daves rule.

The amp actually was very clean inside. It's a closed cabinet, dust can not easily get in. The heat sinks are on the outside. The PSU on the other hand, although without big vents, was quite dusty inside after 30 years.

The PSU works like this in cooperation with the amp:
A small transformer inside the PSU is always on, delivering 24 volts dc via the umbilical cord to the amp.
The power switch on the amp just routes the 24 volts back to the PSU, supplying the circuitry. This then pulls a relay that gives mains to the four main transformers, charging the caps inside the PSU and the amp (157,000uF) through an inrush limiting resistor (6 ohm/20 watts).
After a short delay a second relay takes the resistors out of the circuit giving full mains to the amp.

It's very cramped inside the PSU and many screws to take out. Without removing the side walls you can't reach the transformer nuts, without taking the transformers out you can't reach the sidewall screws.
Well, in the end I managed, already starting to think how to reassemble later on. All brass screws and nuts, nonmagnetic. Blue-tack and long screwdrivers did the job in the end, but don't ask me how many times I lost something in the depths. And you just don't flip a 50kg supply on the head to shake a lost screw out again.

Meanwhile my room filled up considerably

I changed the two caps smoothing the 24 volt standby dc on the PSU board. I inspected the solder joints and found two of the cable screw sockets had broken their connections.

Although the Emitter is assembled quite nicely and things fit well, I think the inrush resistor in the picture below looks a bit bodged. Seems the 10 Ohm/ 20 watt metal case was not up to the job.

Cleaning and reassembling went uneventful. I cut all transformer leads (first hint!) to make fresh connections and gave the screw sockets a bit of Deoxit. After some search in the material I got from ASR I found the right 30 pin connector describing what pins carry what voltages.
I tested with dim bulb tester and variac without the amp connected. As expected, no big current draw from the standby transformer and the 24 volts were present. Good start.
I then proceeded to toggle the main transformers in. Very bright dim bulb, and it did not dim down. I found a short due to wiring mistake by my side. No damage done.
Next try. Click.. bright light.. dimming down... second click... no smoke.

24 Volts still present and left and right channel +- 80 volts present, too (Each channel has it's own voltages). Not bad.

But there should be a secondary winding on each transformer delivering 96 volts dc to the FET drivers. Nothing, all four pins on the connector measure 0 volts. In fact, all four ac voltages directly from the transformer measured on the screw terminals (second hint!) show 0 volts.
Broke my head a while about that. All four windings blown would be bad news. Asked Mr. Schaefer if he had an idea and if my assumptions about the cabling were right.
Before he came back to me I realized that those windings are solid cables while all the others are stranded. So, the 96 volt wires are coming directly from the transformer and had obviously been coated, enamel I think. I needed to burn and scrape it of.

Then everything worked as expected.

Yes, the 99 volts are very close to what the 100 volt caps can handle, but it's without load, so should be ok.
I think the 230 volts we run here now and the amp made when it was 220 contribute, too.


The parts for the amp are in. Next post will be the actual amp repair and then I will close it with some tests.

 

[Rytikar]

The Repair, Part 1:

Ok, parts are in. Resistors, Caps, Diodes and FETs. All from Mouser.

I replaced all 20 gate resistors with good quality 100ohm/0.25watt types. The ones from the good right channel mostly for optics.

I replaced all four 221kohm resistors and the accompanying zener diodes for the bias.

I replaced all electrolytics but the main filter caps with Panasonic FCs or Nichicons. Most of the 470uF/100V had been axial but I chose radial. The cost difference is quite significant and by stretching the leads out they fit well. A bit of silicone for stabilizing. I measured the old ones for capacity and ESR with my LCR bridge. None of them were totaly bad, but all had lost in capacity, some about 30%. All FRACO, btw, made in W. Germany.

I went with IRFP240/9240 FETs, so twenty 0.22ohm/5watt source resistors needed to be added to prevent thermal runaway. ASR has a repair guide were they recommend to isolate the source pin of every FET by drilling in the board somehow and mount the resistors under the PCB. After taking a long look and quite some thinking I decided not to butcher the board. I practiced with old components and found a 'folding' method adding the resistors upright on top, one lead in the original source pin hole. Then bending the source lead of the FET upwards and cutting it so a stump was left over that was thoroughly soldered to the other lead of the resistor, giving stability at the same time.
Took some time but in the end I think it went very well and doesn't look too alien.

I replaced the driver and feedback transistors also located on the heat sink with new ones for good measure.

Some of the parts were located under the copper bars screwed to the underside. Major pain and slow work to get them changed.

That's how it looks now:

When all was done the underside with the copper bars and the filter capacitors was reassembled and the amp was ready to test.

Now, ASR recommends testing with a +-30 Volt regulated power supply at 2 amps. I only have a +-15 Volt supply, albeit with 3 amps.
I also have two standalone 50 volt supplies, but they work with buck/boost regulators. Months before I had tried what will happen if I connect them so I would get +-50 volts out of them. Well, FET blown on one supply. Cannot be done.

So I wired the previously good right channel up with +-15 volts and gave it half an amp.
I was measuring bias voltage to see if something springs to life... Supply got current limited and I got 20mv of bias (10-14 is recommended under normal conditions).
I gave it a bit more current, bias went straight up. No good.

On to the previously damaged left channel. This one took 470mA and had a stable bias at 11mV. Some of the LEDs started very faint to glow.
That looked more promising and I took the chance to see if I could adjust the bias all the way down for later tests with more voltage. Yes, no problem. Very nice. At least no dead short and the amp reacted as expected.

Looking at the weird behaving left side again I finally realized the very embarrassing problem. I had installed the other channels FETs first, amp standing on its side so I could work low and gravity not making all parts fall down. Then I had flipped the amp to stand on the other heat sink to fit in the other set of FETs. I had lost orientation and although I had tried to be as meticulous as possible I had managed to switch the N and P channel FETs around.
Inhaled, exhaled, went down, got some coffee and a walk with the dog. Ok, two hours more work needed, but no damage done and the problem found.

FETs are switched, power applied again... No bias at all... Supply current limiting at one amp, amp happily consuming 30 watts... somewhere.
Half an hour of measuring and looking around I was contemplating over the amp with the power on, when the problem showed itself after about 5 minutes.
With a bang one of the 470uF/100volt caps opened its vents and spew electrolyte over the place. And it was hot! And it was soldered in the wrong way. Yes, there is no silkscreen on the board, but again, I knew the importance and had been very careful. Very embarrassing again, I need to think about me documenting better next time. On the other hand, I've never seen a cap explode live. Quite some spectacle.
And Murphy chimed in for my dumbness. The exploded cap was the worst of all to replace, right under the copper bar.

Cleaned the board, soldered new cap in, supplied power. Channel comes on as the other one, 470mA, stable bias that I also wound down. Phew!

The next day, and I'm preparing for the next step. Variac, dim bulb tester, mains current reader, voltmeters on the bias. Plugged the Emitter power supply into mains and the umbilical cord into the amp.
I knew running it with low voltage would not work as the standby supply would not deliver enough voltage to activate the relays. So I gave it 150 volts.
Turned the power knob on the Emitter to the one position (it has two). Nothing.
200 volts... The dim bulb glows bright, the inrush relay starts clicking in and out. Not enough juice through the bulb or a short somewhere.

Problem now was, the DIY dimbulb tester that had served me for years could be set to a 60 watt bulb or a 100 watt bulb. Good enough for everything I had worked on as of now. Not enough for the Emitter, by far. The power supply ALONE takes 120 watts idle without the amp connected. So I modified my dim bulb tester to switch two 100 watt bulbs parallel. Thats the best I could do.
Inrush relay clicked in, but not the relay short-cutting the inrush resistors. Not enough juice. But the amp got at least some power through the inrush resistors. The LEDs started to glow more and the bias on both channels got up to 7mv.
Considering my options for a day I decided to give it a try.

Full 230 volts line voltage, no dim bulbs in series.
Inrush clicked in, two seconds later inrush shortcut clicked in, two seconds later speaker protection in the amp clicked in. No smoke, no explosion, 17mv of bias and lots of yellow, green and red LEDs glowing (the reds show what input is selected, not an error).

So far so good:

I will close part one here, because we're not done, sadly.
Three problems remain and I've emailed with Mr. Schaefer again who is patently helping where he can. Absolutely amazing support.

1. Bias is at 14-15mv (it fell a bit after 15 minutes) but at lowest setting. That's a bit too high and the bias zeners may need to be adjusted.

2. I gave the amp some signal testing on power level one and it actually worked quite fine (up to ca. 100 watt). When I switched over to power level two the amp produced quite astonishing 130 watts with 100mv input only. Then something started to smoke, I think the 0.22 ohm source resistors. Probably oscillation. May need higher value gate resistors.

3. Not mentioned above but restricting my testing quite severly is the fact that two out of three times when connecting the power supply to mains or when that was successful switching the amp on the RCD breaker in my house trips. Something is not ok in the power supply.

We'll see about that in part two. So close...

 

[Rytikar]

The Repair, Part 2:

At the end of the last post we had a working amplifier with three problems left to take care of.

The most urgent was the tripping of the RCD breaker nearly every time I plugged the power supply in. So I opened it up again and started measuring. I found that between ground and life was a resistance of 'only' 43 ohms. That seemed not right and after some research I was sure that was the problem.
I started taking the PSU apart again (third time, getting good at it) by removing one transformer connection after the other from the screw terminals. Even after all was disconnected the resistance hadn't changed. I took the board out and after some poking around I found out that the metal screw securing the inrush limiting aluminium cased resistor was causing the bridge. Once removed no resistance could be measured. I replaced the one metal screw with two nylon screws, fixing the problem. While I was at it I replaced the 'how you're doing' parallel construct with two 25 watt cement resistors in series (see picture in post above) with one 25 ohm/50 watt aluminium resistor. The original value was 2 times 8.2 ohms the available replacement has 25 ohm, giving 7.1 ohm instead of 6.2. Good enough for Denmark.

I used heat conducting compound to glue them together:

Next was the bias being too high with the adjustment completely turned down. With some help from Mr. Schaefer I realized the bias voltage was created by an 'adjustable zener' constructed mainly by a transistor and a collector-base resistor adjusting the voltage. The collector-base resistor was one 1k27 resistor and a 200 ohm pot in series. Now, the new FETs are reacting more aggressive on gate voltage changes than the old ones so I had to be careful changing values. I added 20k in parallel to the 1k27, then 10k, which proved to be a good value. With no good standard value available and not wanting to disassemble the Emitter again to reach the underside I opted for soldering the 10k resistors into place in parallel to the 1k27. I then adjusted the bias to 12mV.

It's more tricky than it looks to reach down there:

A little side quest was finding the reason for a seemingly misaligned bias for the P and N channel FETs of the left channel. There are four bias measuring points, one for each P and N side of each channel. Three measured 12mv, one 10.5mV.
Consulting Mr. Schaefer again and in the end measuring the source current for each FET (conveniently using the new 0.22ohm resistors) it showed that everything was ok with a nicely balanced combined load. Finally, I reduced the bias to 10mV.

source resistance | source voltage in mV | source current. Yes, the right channel takes bit more but is balanced. Yes, that's 2.227 amps of idle current.

I could not reproduce the 'smoking source resistor' phenomenon but I think I have an explanation. I used my 'Analog Discovery' (https://analogdiscovery.com/) device together with TheStuffMade's (https://www.youtube.com/channel/UCS8pMGUEvhAPMt9_ml5iyRg) audio test suite for testing. I had the Emitter on Level 1 with the volume cranked all up, playing with the input value to generate different output levels.
The test suite has a problem with decimal numbers for us here in Europe. If you forget typing 0.200V for value the AD should output but type the European style 0,200V instead, the suite will convert the input string to 200 and the Analog Discovery will output all it has to give. That's an input level of about 3.6 volt rms to the Emitter.
While testing I happily switched to Level 2 without reducing the volume level.
I think the Emitter was outputting all it's modern FETs and the power supply could give. And that's a lot. And that may have been a bit too much even for the 5 watt source resistors (Mr. Schaefer said 2 watt will do, but recommended 5 watt).

The Emitter is done cleaning, reassembled and tested. I have some graphs I will show in the next and final part very soon.

 

[Rytikar]

Wrap Up:

It's finally time to close this repair and restoration thread from my side.
The Emitter has been up and running and I've listened to it for several hours during the last week. It works fine without any problems and sounds wonderful.
I actually had the feeling it got better with a little break in. Don't know if it's because it has not been used for some years or all the new caps needed to reform a bit.

ASR in their information of the several Emitters during time describes the old ones, before 1991 when a total redesign of the circuit happened, as sort of slow, warm and tube-like. That actually fits nicely with my latest adventures into tube amplifying with a home build KT88 single ended tube amplifier (

).
It matches my current listening preferences quite well and I will swap between the Emitter and the KT88 once in a while.

Before setting the amp up I made some measurements I want to share with you. Some of them make me wonder a bit but it may well be that the new test setup I use is not fully understood yet. As of now my ears are pleased with what they hear.

Tests have mostly been done on power level 1, as my test setup is not able to handle the power the Emitter can produce.
It measured up to 100 watts without breaking a sweat and I risked for some seconds to push it. It needs some input level, but at 2 volts rms input on power level 2 I got it to produce about 53 volts rms output without clipping before turning down again. On 8 ohms load that's 350 watts per channel.

Let's start with frequency response from 10Hz to 50kHz at 1 volt rms (ca. 10 watts).
One may think the two channels are not matching, but the actual difference is minor when looking at the dBr scale.

Next is THD vs frequency at 1 watt.
Well, it's ok but tubelike it is, although I may need to retest this one once I am willing to move the 100kg over again.

THD vs power from 10mWatt to 100 watt.
Looks a bit funny to me, going down with power.

Last but not least a spectrum diagram at 10 watts.
Some harmonics are present but at under 50db. Need to read more about that.

That's how the Emitter is set up now. My room is small and an obvious question arose. Can an Emitter be too big for a room... Nah... But you for sure can feel the 400 watts idle power warming up the 8 square meters.

And yes, I got myself a clear acrylic top.

Thanks for reading and following along.

Btw, I spent about $70 for parts and some 30 hours work into the Emitter. Well worth it, I think.

Special thanks to Mr. Schaefer from ASR. He was invaluable to give insight into some questions I had about the Emitter and the PSU. And without his rebuild guides I would not have been able to finish this. Amazing support for a product I bought 30 years ago.

 

[Sound Dragon]

I enjoyed your progress bud. Enjoy that fine amp in good health!

 

[Dazaa]

Very interesting thread. Fwiw the original fets are still available here in the UK from a small reputable supplier at £15 each

 

[Rytikar]

Just a small update. I found the problem with my Analog Discovery setup. The graphs shown above are misleading and not representative. When I get the time to move the Emitter to the bench again I will repeat the measurements and update the graphs.

 

[Rytikar]

One more update.
Over time it showed that the Emitter was humming a bit, sometimes more, than going away while playing, never vanishing. I decided to swap the twelve 10,000uF capacitors in the amplifier unit (date coded 89/12, btw). Because of the way they are mounted it was necessary to get the exact size and they needed to have screw terminals. Got myself twelve 12,000uF Nippon Chemi-Con. Swap was easy, no soldering, just screws and handling of big copper bars. Still, took two hours.
Worked as hoped, hum is gone and the sound improved nicely, especially in the bass. Don't know what it is that makes this amount of power reservoir noticeable even on lower listening volumes, but it is. Maybe Mr. Schäfer was on to something with his power supply design.
When I have the time I will swap the eight 4,700uF capacitors in the power supply section, too. The are just snap in and are not as expensive as the screw-ons. The amount of work needed to get to them is a bit distracting, on the other hand.

Looks great, though, I have to admit.

 

[Bumpalump]

Im working on exactly same amplifier, burned right side

 

[Rytikar]

Hi, I lookd at you're video, a bit hard to understand. I hope you get it fixed!

 

[Bumpalump]

Not yet, i still have +65V at gate resistors on both P and N side, all the supply voltages are correct, all the parts are ok, one outpit pair installed