I have had a Hafler XL-600 amplifier in my system for about a year now. I bought it off of Craigslist from a local seller who was the original owner of this amp. It is super clean and is about as close to new as you can expect from a 20+ year old amplifier. The only thing it really needed once I got it home aside from cleaning the fan blades off was to adjust the output stage bias. However, being an avid DIY’er I can never leave well enough alone.
I know that the later generation 9300/9500 amplifiers were very well regarded and highly reviewed back in the 90’s. Not that the XL series was any slouch, but they didn’t seem to have the same reputation as the later gen 9300/9500 amps. I started wondering what the differences were between the older XL series versus the newer non-balanced input 9300/9500 Hafler amplifiers. After examining the schematics [which are attached] for the XL-600 and the newer 9300/9500 it looks as though the main change was a simplification of the predriver circuit. The later amps gained generic 3 pin 24V regulators to the DC rails and the elimination of several parts including the cascode transistors (did I identify that correctly?). This offered “regulated” DC rails in the later 9300/9500 amps.
After thinking about it for a while it seems it would be reasonably simple to construct a PC 40 replacement board based on the later 9300/9500 predriver and marry it to the XL-280 or XL-600’s output stage. In fact I’m sure it could be used for the DH200/220 output stage as well although hI have not investigated it. From the schematics it appears that the output stage between the earlier XL series and later 9300/9500 generations of Haflers are identical except for the fact the DH and XL amplifiers used the more rugged metal can TO-3 devices, which I much prefer. I know that the XL series input differential SK163/SJ74 JFET’s have been obsolete for a long time, but the original JFET’s could be carefully removed from the working original PC 40 boards and installed into a new board. Not to mention that there are still a few sources for matched 2SK170BL/2SJ74BL devices albeit at a premium.
Aside from the JFET’s the other components in the 9300/9500 predriver all look to be easily sourced and cheap. I was thinking that a simple zener + emitter follower regulator could be integrated onto each board so each board would have its own dedicated +/- regulator. The zener + emitter follower regulator would also have a much lower noise floor than the standard 3 pin variety used in the 9300/9500.
I know based on other threads here that others have modified and changed the front end of the old Hafler amps over the years, but I haven’t seen anyone that has done this type of upgrade. Musical Concepts offers their PA-4 kit, but at a very hefty price. I think that these updated PC-40's could be built at a cost of < $100 even with having to go and pay a premium for the input JFET’s. Without the cost of the JFET’s I would think these could be built very cheaply at < $50.
Now some are probably wondering “Why not just sell the XL-600 and buy a 9500?” So I’ll go ahead and answer that question.
I like to mod my electronics.
I like the looks of my XL series better.
I like the metal can TO-3’s better than the plastic TO-3P’s.
I like the fan cooled heatsink better. Many would say the later 9500 is the better amp because of this, but the fan affords much better cooling. I am running 575 mA of output bias and can hold my hand on the outputs no problem and at this bias level the fan barely speeds up. Sitting in my chair the fan is inaudible when music is playing and it is barely audible with no music playing. It is my understanding that the 9500 cannot get much past 400 mA without getting too hot. If I’m wrong on this last point let me know!
One thing that I do not like is the reports I have seen of overheating on the 9300/9500 boards around the output mosfets and trace failures as a result. This seems like some marginal engineering was at work here. Due to this fact I would be quite leery of buying one without being able to inspect the main board.
Just wondering what others thoughts are on this possible upgrade for the PC 40? Would it be worthwhile? If it is a worthwhile upgrade would anyone be interested in some bare boards if a small batch were to be built and sold at cost?
I know that the later generation 9300/9500 amplifiers were very well regarded and highly reviewed back in the 90’s. Not that the XL series was any slouch, but they didn’t seem to have the same reputation as the later gen 9300/9500 amps. I started wondering what the differences were between the older XL series versus the newer non-balanced input 9300/9500 Hafler amplifiers. After examining the schematics [which are attached] for the XL-600 and the newer 9300/9500 it looks as though the main change was a simplification of the predriver circuit. The later amps gained generic 3 pin 24V regulators to the DC rails and the elimination of several parts including the cascode transistors (did I identify that correctly?). This offered “regulated” DC rails in the later 9300/9500 amps.
After thinking about it for a while it seems it would be reasonably simple to construct a PC 40 replacement board based on the later 9300/9500 predriver and marry it to the XL-280 or XL-600’s output stage. In fact I’m sure it could be used for the DH200/220 output stage as well although hI have not investigated it. From the schematics it appears that the output stage between the earlier XL series and later 9300/9500 generations of Haflers are identical except for the fact the DH and XL amplifiers used the more rugged metal can TO-3 devices, which I much prefer. I know that the XL series input differential SK163/SJ74 JFET’s have been obsolete for a long time, but the original JFET’s could be carefully removed from the working original PC 40 boards and installed into a new board. Not to mention that there are still a few sources for matched 2SK170BL/2SJ74BL devices albeit at a premium.
Aside from the JFET’s the other components in the 9300/9500 predriver all look to be easily sourced and cheap. I was thinking that a simple zener + emitter follower regulator could be integrated onto each board so each board would have its own dedicated +/- regulator. The zener + emitter follower regulator would also have a much lower noise floor than the standard 3 pin variety used in the 9300/9500.
I know based on other threads here that others have modified and changed the front end of the old Hafler amps over the years, but I haven’t seen anyone that has done this type of upgrade. Musical Concepts offers their PA-4 kit, but at a very hefty price. I think that these updated PC-40's could be built at a cost of < $100 even with having to go and pay a premium for the input JFET’s. Without the cost of the JFET’s I would think these could be built very cheaply at < $50.
Now some are probably wondering “Why not just sell the XL-600 and buy a 9500?” So I’ll go ahead and answer that question.
I like to mod my electronics.
I like the looks of my XL series better.
I like the metal can TO-3’s better than the plastic TO-3P’s.
I like the fan cooled heatsink better. Many would say the later 9500 is the better amp because of this, but the fan affords much better cooling. I am running 575 mA of output bias and can hold my hand on the outputs no problem and at this bias level the fan barely speeds up. Sitting in my chair the fan is inaudible when music is playing and it is barely audible with no music playing. It is my understanding that the 9500 cannot get much past 400 mA without getting too hot. If I’m wrong on this last point let me know!
One thing that I do not like is the reports I have seen of overheating on the 9300/9500 boards around the output mosfets and trace failures as a result. This seems like some marginal engineering was at work here. Due to this fact I would be quite leery of buying one without being able to inspect the main board.
Just wondering what others thoughts are on this possible upgrade for the PC 40? Would it be worthwhile? If it is a worthwhile upgrade would anyone be interested in some bare boards if a small batch were to be built and sold at cost?
Attachments
Bit of thread jacking here ....
Given the opportunity and while many Hafler experts may look at this i wonder if musical concepts PA4 boards is an option for the XL 600
Farther more i have just installed PA2 boards in XL 600 and i am runing into troubles so i need schematics and help
Kind regards
Sakis
Given the opportunity and while many Hafler experts may look at this i wonder if musical concepts PA4 boards is an option for the XL 600
Farther more i have just installed PA2 boards in XL 600 and i am runing into troubles so i need schematics and help
Kind regards
Sakis
Sakis,
The Musical Concepts PA-4 boards are an option for all of the old DH and XL series Haflers, but an expensive one IMO. I mentioned this about half-way through my original post. I'm a bit of a cheap skate when it comes to this kind of thing. I would rather come up with my own upgrade solution and avoid the $$$ where possible. 😀
The schematics are readily available from Rockford Fosgate. I called their tech support and within 5 minutes the schematics were in my email. While I had them on the line I went ahead and had them send me the Hafler Iris preamp schematics as well.
sorry to hump into your thread asking for info ...
yes Rockford people been the same helpful to me in more than one occasion In my case the original boards work as is with no probs
problem is that costumer has already the PA2 ( not ! PA 4 ) boards and wants them in ... so gate capacitors and implementation is one issue , bias behavior is the other plus it seems that one of the boards has some trouble as described here :
http://www.diyaudio.com/forums/solid-state/239454-hafler-xl-600-pa-4-pa-2-boards.html
Thanks for the info so far ...
Kind regards
Sakis
yes Rockford people been the same helpful to me in more than one occasion In my case the original boards work as is with no probs
problem is that costumer has already the PA2 ( not ! PA 4 ) boards and wants them in ... so gate capacitors and implementation is one issue , bias behavior is the other plus it seems that one of the boards has some trouble as described here :
http://www.diyaudio.com/forums/solid-state/239454-hafler-xl-600-pa-4-pa-2-boards.html
Thanks for the info so far ...
Kind regards
Sakis
Good point!
Options,
1) A new 25 VA toroidal is $20 from Mouser.
2) A higher voltage rated transistor (i.e. 2SC5171) could be used in the zener emitter follower to drop the 100V rail down to 24V. The CB voltage drop would be around 75V so a resistor of around 5W would be needed in that location which would eat up some board space. Plus decent heatsinks would be needed on the transistors.
In the end, option 1 would be the better solution even though it is more expensive. Isn't that the way it always goes, the more expensive solution is the better one!
"1) A new 25 VA toroidal is $20 from Mouser. "
You're building a 10W amplifier?
You only need one transformer for how ever many channels you need for the front end, you need a separate secondary for each output stage channel.
You're building a 10W amplifier?
You only need one transformer for how ever many channels you need for the front end, you need a separate secondary for each output stage channel.
i understand the need to tinker with it.
i suffer from that disease myself.
🙂
isn't the XL600 the high power version of the XL280?
if so, i'm thinking that the XL600 design front end may a better performer overall than that of the transnova design.
mlloyd1
i suffer from that disease myself.
🙂
isn't the XL600 the high power version of the XL280?
if so, i'm thinking that the XL600 design front end may a better performer overall than that of the transnova design.
mlloyd1
MLLOYD1,
Yes that is the one. Same PC40 front end as the XL280 and 4 pairs of Hitachi outputs per channel with 25 volt higher rails. Basically it was what evolved out of the DH500.
Well I certainly would not want to take a step back or even laterally in performance.
I recently did some upgrade work on a Hafler SE100 and there was a lot of improvement to be had by modifying that design. Economic constraints greatly affected the parts used in the original manufacture of that unit. With the XL600 I may just need to look at ways to improve what is already there. I'll have to investigate some more...
Yes that is the one. Same PC40 front end as the XL280 and 4 pairs of Hitachi outputs per channel with 25 volt higher rails. Basically it was what evolved out of the DH500.
Well I certainly would not want to take a step back or even laterally in performance.
I recently did some upgrade work on a Hafler SE100 and there was a lot of improvement to be had by modifying that design. Economic constraints greatly affected the parts used in the original manufacture of that unit. With the XL600 I may just need to look at ways to improve what is already there. I'll have to investigate some more...
Last edited:
Not only that but the XL600 and the XL280 are both stable down to 1 ohm, none of the transnova amplifiers are, also I have owned several hafler amplifiers... DH220, DH500, P230, P7000 XL280. The XL600 is the holy grail of all the above, not just the power as I am listing to some great tunes at maybe 1 watt as we speak. The overall musical presentation of the XL600 is quite breath taking, the overall phase coherency throughout the entire bandwidth is to say the least phonemail! Using the TAD TD2002 drivers will reveal any weakness in any given power amplifier... this amplifier is so smooth as in easy on the ears and so warm and musical its so nice too listen to, its like the electronics left the room and the music stayed behind. The transnova series has a slightly better damping factor, yet for overall musicality the XL600 is to date the best sounding Hafler amplifier i have had the pleasure to audition in my system. 5 Star
One more note on the XL600 verses the 9500 or 9505, the XL600 still in most all respects from stability to listenability the XL600 supersedes all other Hafler amplifiers. Also if you look at any transnova amplifiers specification sheet you will not see any IM or TIM ratings, yet the XL series are published. I do like the transnova but not nearly as much as the XL series. When the music matters as it always should, the XL600 wins at least in my system hands down.
TAD MAN;
i know what you mean. my main listening amp is a borbely Servo 50. topologically, very similar to hafler xl-280/xl-600 designs. in fact, i have liked it so much over the years that it is the only amp i have ever made that i have not disassembled to use the parts in another project.
mlloyd1
i know what you mean. my main listening amp is a borbely Servo 50. topologically, very similar to hafler xl-280/xl-600 designs. in fact, i have liked it so much over the years that it is the only amp i have ever made that i have not disassembled to use the parts in another project.
mlloyd1
Sorry I missed the followup on this thread.
I have owned a few of the old Hafler amps as well, the DH200, DH220, & XL280 and I will say that so far the XL600 is far and away my favorite Hafler amp. I have never owned any of the later Jim Strickland Transnova series though. I did listen to a 9500 at length in a local hifi dealers showroom back in the early 90's though.
I have owned a few of the old Hafler amps as well, the DH200, DH220, & XL280 and I will say that so far the XL600 is far and away my favorite Hafler amp. I have never owned any of the later Jim Strickland Transnova series though. I did listen to a 9500 at length in a local hifi dealers showroom back in the early 90's though.
- Status
- Not open for further replies.