POOGED Hafler transistor blowup

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
The way this is done is maybe 3 to 9 wires of each polarity @24 gauge (Cat5) are braided together, to cancel out noise and impedance effects. This becomes equivalent to 18 to 14 gauge wire, but as each wire is individually insulated, you have more surface effect to transmit high frequencies, and less interference, particularly if you braid the wires; and plenty of wire to carry current. At least that's the theory. I was going to use a 3 wire-pair braid from the inputs to the PCB (3+, 3-) and 6 or 9 pairs to the binding posts. I wasn't going to use this with the power supply wiring (though the AA article suggests something like it). Sometimes magnet-winding wire is used instead for this. The thinness is not supposed to be a problem, because there are multiple leads just like in twisted primary wire--but each strand is insulated, and the braiding is supposed to reduce interference and noise.

Is it totally wacko?

Anyway, yes, I am going to do the MOSFET repair first.
 
Silpads are still for convenience. There's no thermal grease required, for serial production that'll be an interesting cost saving. There's even a silicon foil type for uneven surfaces, called GEL foil. 1/8" thick and compressable to 1/16" without glassfiber reinforcement but has poor thermal resistance. I've used some of the early silicon foil for Japanese devices with uncommon dimensions i wasn't able to get micas for, 20 years later i can but they're filthy expensive.
Nelson Pass posted some time ago that mica and thermal grease is still the ticket. The unusual white ceramic Al2-O2 flakes from Aavid Thermalloy i'm using on my Krell toaster are better than mica. Composed of 3 pieces of Aluminium Oxyde under each of the same type Japanese output devices i mentioned it renders better heat transfer than mica, a single ceramic insulator would do even better.

Personally i loathe spaghetti in amplifier cases. Neatly folded and bundled hookup wires with wire straps is nice, but braided ones have a simular efect on me as the Playboy centerfold. Make it different color braiding and you have to post a pic for other people's personal image collection. I favor teflon sleeved silver coated stranded copper hookup wire, currently waiting for several hundred feet in diff. colors and AWG sizes to arrive from ApexJr.
 
Administrator
Joined 2004
Paid Member
Hi Michael,
Is it totally wacko?
Yes. And braided hoses make your car go faster! ;)

A fine strand copper cable is going to work the best, that's if the effect can be heard at all. I work with CAT-5 cable every day almost. There isn't a shred of it in any audio thing I own. I have lot's in the garage.

CAT-5 does work well for netwoking, and great for telephone wiring.

I don't care for Silpads at all, especially the compressable ones. At least mica and grease are a known quantity and they work well. In production I might use some kind of silpad type product. You can't trust workers to do a proper job these days. It's not uncommon to find crushed solder balls under transistors from the factory. Nice eh?

-Chris
 
Michael,

The comments by Jacco and Anatech are spot on. The original article on how to POOGE the DH-200 was OK in its day. There are a few good points made in it but some of the mods it proposed were "over the top."

Look carefully at changes made between the DH-200 and DH-220. Also, look closely at the differences in parts lists between the DH-120 and the Hafler 9130 to see what I mean. Polypropylene caps in certain spots are better than mylar, metal film resistors are better than carbon resistors in many spots. But, neither will make a hearable difference when used in all spots. Also, there is a lot of snake oil sold to justify high prices for certain types of wire. I would not slavishly follow every part of the POOGE article. Plus, like it or not, there are some circuits that sound better than others and the DH-200 circuit has seen its heyday.

Also, the DH-200 circuit was laid down in the "boom and tizz" era of HiFi progress. Speakers, woofers and tweeters, have since improved to the point a more even reproduction is desirable. The DH-200 was OK for its time and still represents a fairly good value, but you can't make a silk purse out of its sows ears. This is why I stopped doing major mods to my DH-200 and installed the PCBs from Musical Concepts.

My $0.02 ;)
 
anatech said:
braided hoses

Highly recommended,
both for American oldfashioned displacement as for Swedish 2 cylinder boat diesels. One day i'll build a poweramp with braided hosing :clown:
Works even great on Turbo Hydraulic automatic trannies :c_flag: =>
 

Attachments

  • braided.jpg
    braided.jpg
    25.5 KB · Views: 247
Okay, okay, I'll lose the Cat-5. Now, what am I going to do with all that extra wire in my tool shed?

It sure is a great value to be able to get the advice of those who post on this forum. Many thanks again to all of you.

Jacco, what AWG of silver-coated wire do you use when wiring amps? Do you use a different size for inputs, output to the binding posts, and power supply voltages?

I might just go to 16 AWG twisted strand Cu primary wire. The original was 18 AWG. But I will look into the silver-coated stuff--it doesn't seem so easy to find.

Dick, bottom line on the Musical Concepts mod-- Does it bring the DH-200 chassis up there with the best of what's available today? What would one I have to buy commercially to equal it?
 
Michael,

MC sells its own branded amps and has one for around $1000 that is essentially what you would have with the MC PCBs and PS caps.

Ask around on the Madisound and Parts Express tech message boards to get opinions other than mine. Ditto over at Audio Asylum. And, do a search here at DIYaudio. The figure of $1000 is often mentioned as what one would have to pay to equal the chassis/transformer/heatsinks/MOSFETs you have with the PCB PS cap mods I recommend.

Plus, you could sell the DH-200 PCBs you now have to help defray some of the cost for the MC kit.

I give my $0.02 but ask around and understand you have to make the final decision :xeye:

Here is the link for teflon covered silver plated wire that is recommended:

http://www.apexjr.com/

18 ga twisted pair is good for signal wiring and 16 ga is sufficient for PS wiring. If you do some re-wring, study how the MC mods do the ground/signal input for their PCB kits with the DH-200 chassis. Also, if you get the MC kit it will come with most of the wiring you need internally for your DH-200 chassis, plus a nice pair of RCA jacks! :)
 

fab

Member
Joined 2004
Paid Member
Dick West said:
Michael,

...
Also, the DH-200 circuit was laid down in the "boom and tizz" era of HiFi progress. Speakers, woofers and tweeters, have since improved to the point a more even reproduction is desirable. The DH-200 was OK for its time and still represents a fairly good value, but you can't make a silk purse out of its sows ears. This is why I stopped doing major mods to my DH-200 and installed the PCBs from Musical Concepts.

My $0.02 ;)

Dick, Ihave not heard the MC circuit and can not comment on the sound and anyway one's appreciation on sound is subjective because of too many factors. I just want to say that from the pcb layout I have figured out most of the MC driver circuit and the schematics is "generally" probably an older concept than the one in the DH-200/220. However, the MC may use better components and pcb though. The latters may affect the sound positively. I suppose that upgrading some resistors and caps in the DH-220/200 could possibly bring it up to the level of the MC (and maybe more?)...... but it is only a supposition.

My $0.02 (CAN)
 
Apex Jr.com-- Wow, what a great site! And what prices--makes you want to buy stuff, even if you don't need it or know what you're going to use it for. Even if you don't even know what it is.

Ok, it looks like we have a pro and a sort of con now on Musical Concepts. It's a discussion! Anyway, Dick, I took your advice. Posted on the Parts Express site, and read all the comments I could find on MC at the audioasylum site (there are plenty). At PE, Dick is the only poster talking about MC. Just trying to make sure you're not a shill for MC, Dick. ;)

I have to say, but for one or two expressing disappointment on Audio Asylum, overall the comments on MC stuff are overwhelmingly positive. I could do the caps upgrades, etc myself fairly easily, but I believe what I am gathering is that John Hillig has simplified the circuit and made the sound more revealing, as well as improving components. I might follow Dick's advice on my as yet unmodded DH-200 and get the MC upgrade.
 
Fab,

I was privileged to spend some time looking at the schematic for the MC PA3-B schematic. My electronics knowledge is not up to your level but the circuit appears to be a current fed long tailed pair (2N5551) followed by pre-drivers (BC550/560) in what looks like a casode connection to drivers (MJE340/350). No input cap. Bypass filter caps are 100uF/100VDC. A very simple circuit with what I am sure are carefully selected and matched parts. The feedback network has what seem to be extra parts and I am sure it is the "heart" of the sound of this circuit and has been carefully titrated by observing o'scope readouts.

I have owned both the PA3A and the PA3-B modded amps and the PA3-B is MUCH better, more refined and smooth, in sound. Someday when I have saved enough pennies I will get the next generation (PA3-C) PCBs from MC ;) .

The Darcy Staggs regulator circuit is a straightforward circuit using LM317T/337T devices with either a 20K 25 turn pot or a #1N4759A zener as a voltage reference bypassed with a 22uF 100v cap. It regulates ~85VDC down to ~63VDC. Staggs jumpered out (on the DH-220) D15/16 and R31/32. He used a TRIAD FP120-100 12W120VAC out @0.1A transformer.

I have used several DH-200/220 amps over the years and some have had varying amounts of mods, even a dual chassis dual mono DH-200. The MC mod PCBs throw a larger soundstage with more discrete instrument locations and a very smooth, detailed, non-fatiguing sound. Time and again I would go back to the MC amp over any combination of Hafler amps available. At first listen the MC amp seems a tad bass shy but then I realize that the MC bass is more natural and goes deeper than that from the Haflers. Also, strings sound more like strings on the MC and percussive transients (drums, cymbals, etc.) seem more dramatic and controlled.

Well, to each his own.

Dick
 
Administrator
Joined 2004
Paid Member
Hi Dick,
My experience is that matching certian parts almost always improves the sound across the board. Replacing of substandard parts (from an audio point of view) improves the sound some more. Just don't get caught up in the capacitor of the week thing.

Board layout has a dramatic effect on sound quality given the same schematic, so there is another point to consider.

-Chris
 
Michael,

Most of the guys at the PE board don't recognize my name. I post there very infrequently. The fellow who did respond has purchased from me 2 Haflers and, for a buddy, one MC amp. He did not like the MC amp, then he did, then he didn't. He does like the Hafler sound for his home theater and I have enjoyed exchanging some messages with him.

However, the MC amp he listened to have the PA3-A version of the PCBs, not the later PA3-B which have an improved sound. I can hardly wait to try out the latest PA3-C PCBs. No, I am not a schill for MC. John Hillig's work is so good he does not need any schills. Satisfied customers take care of this function, plus he has been in business for over 20 years and has gained an excellent reputation amongst discerning listeners.

A post on the Madisound board (without my name) might get a better selection of responses about MC products. I watch messages on the PE board on how to build a woofer but I go to the Madisound board for info on how to build a speaker sytem, crossover, and electronics. These two message boards do have a slightly different audience. Many posters at the MAD board are engineers or actually deisgn and sell speakers for a living. They are a high level bunch! If I have a Hafler amp to sell I post a message at PE but not at MAD

Cheers...........

Dick
 

fab

Member
Joined 2004
Paid Member
Dick West said:
Fab,

...
Well, to each his own.

Dick

Dick and MichealIP, do not get me wrong , I have nothing against MC since I have not "heard" it so how could I prefer Hafler or something else. I just wanted to clear out that one can not try to qualify the sound of a driver board based on the date of design of a circuit. A lot of new amps still have the same basic design as in 1970. Maybe today, there are better parts available - if they are really used in the circuit. The MC board appears at first sight of a better quality (at least the pcb) than the original DH-200/220 pcb. But again, as already stated by many over this forum, I think that the only way to judge the sound of a circuit is really to hear it by yourself in your own system with your loved music. Iknow that Dick you have done a comparison in sound with modded Halfer amps and the MC board version B and you prefer it. This is a good info but one need to realise that there are a lot of possible mods to be done on the Halfer DH-220/220 amps since its modular design. Have you tried all possibilities?

I think MichealIP has the possibility and the chance to at least compared one "specific" modded DH-220 amp with another one with MC board so he can judge by himself which is good for him.
On my side, maybe I should have done it since I had 2 DH-200 amps but now I gave one to my friend...
 

fab

Member
Joined 2004
Paid Member
Dick West said:
Fab,

The Darcy Staggs regulator circuit is a straightforward circuit using LM317T/337T devices with either a 20K 25 turn pot or a #1N4759A zener as a voltage reference bypassed with a 22uF 100v cap. It regulates ~85VDC down to ~63VDC. Staggs jumpered out (on the DH-220) D15/16 and R31/32. He used a TRIAD FP120-100 12W120VAC out @0.1A transformer...

Dick

Thank you Dick for this info related to post on the other DH-200/220 thread http://www.diyaudio.com/forums/showthread.php?postid=956902#post956902

The Stagg's circuit is then very similar to the one I have suggested in the above thread.

Maybe mmerig could use the MC version B board with a regulated power supply for the driver and get even better results !?...
 
Fab,

My main comparison between MC and DH was with a DH-200 that had all electrolytics bypassed with polyprops, 40K uF in a separate PS for each channel. The PCBs were run with a regulated 60 VDC and the output stage with 55 VDC. The PS transformers were 6 amp old Dynacos. All wiring was done with good stuff and it had gold inputs and outputs. Each channel was on a separate chassis with its own independant PS.

When I got all the voltages settled I fired up the 2 channels to a good stereo CD, like Jazz in the Pawnshop. WOW! was my first response. The sound stage was wide and huge and each instrument seemed to be playing in its own space. There was no hesitation to reproduce any note at any frequency and complex passages were resolved better than anything I had before heard.

My speakers are the Olsher PolyNatalia with top notch Audax and Morel drivers, etc., etc., in a moderately live 22' X 24' room.

I thought "This is it! This is sonic Nirvana!" Then I gave it all a good comparison with the DH-200 chassis that had the MC PA-3B cards and 52K uF PS caps (new and fresh ones). The MP amp sounded a little different from the tricked out Hafler. But, soundstage, instrument placement, passage resolution, sonic integrity, all seemed to be equal or superior to the dual mono regulated front end Hafler. Plus, for some reason, the MS just seemed easier to listen to for extended periods of time.

I have no signal generator or o'scope or other complicated instrumentation. My decisions are made on how things sound to me.

So, I concluded that life would be better and simpler to use the MC amp rather than the dual chassis Hafler with all the extra caps and wiring and front end regulation, etc. I made this decision and have never looked back.

Looking forward I might try splitting my MC amp onto two separate chasses each with its own PS with 52K uF in caps. But, for now, that will be a Winter's project. I know of a fellow in Toronto that has a dual mono MC amp, complete with hexfred diodes, etc. If you are ever in the Toronto area I could inquire as to whether he would welcome a visit from an obviously avid audiophile.

Later..........

Dick
 
Dick, you do know, I hope, I was pulling your leg on that shill thing. I never got the feeling you were doing more than give your honest assessment, but the detailed explanation of the amps you were comparing and the history on those mods was very helpful in filling out the picture. And I agree with Chris. With a scope and a signal generator, golly, I'll bet you could give at least a nickel's worth in every post.

Anyway, I have posted for other opinions on a few other sites, but I am not hearing much back yet. I get the feeling that you may be the only one in the world besides John Hillig who has compared different Hafler mods so carefully. I had some interaction with Hillig in the past, so I have considered the PC-3C upgrade already, but I think I am just about ready to go ahead and plunk down my cash at this point. I'm going over his web site again now.

I notice that on the "Signature" and "Platinum" versions, which bring a premium price, one thing that is mentioned is "additional power supply bypass capacitors for superior resolution". Dick, can I presume that these are more or less equivalent to the PS bypass caps in the AA POOGE, and something I can add easily myself?

On the big PS caps, I already invested in a pair of new Cornell-Dublier @ 22K uF each, and found they give a nice solid bass with even my stock DH-200, or the modded one, on a single chassis. I don't know how far above the original 10K X 2 total it is safe to go without upgrading the bridge rectifier, but I had an inquiry to Mr Hillig a few months back about that, and he seemed to feel up to 55K or so is no problem, as the stock rectifier can handle it. The POOGED DH-200 I have has two 18K caps. I don't know if it means anything, but these are much bigger than stock, about 2 1/2" wide by 4 1/2" tall (so you have to leave the cover off to use them) and they have the name "Kellner E.I." on them. They look like blue CDs made for Kellner, but I haven't figured out if there is supposed to be anything special about them other than the 18K, like maybe a lower ESR? They sound about as good as the 22Ks. (My 22Ks are same size as stock 10Ks.)
 
Michael,

As you before have been told, you are too impatient. 3-4 days turnaround for a post and reply are more the norm.

30K uF PS cap is a figure often stated before the power switch and bridge go up in smoke. I picked up several Cornel Dubilier time delay relays (pulls) with a 120VAC field coil. With a simple resistor their time delay can be set from a few seconds to a few minutes. I use one with a 5 second delay to bypass a 50 ohm 50 watt power resistor. That resistor, in series with the incoming AC for those few seconds let the big caps charge gracefully and no lights dim or switches blow up. I would not be without a soft start for the big caps.

There is some snake oil spread around about caps. The ones Hillig sells for $80 the pair are excellent, that much I know. I have an extra pair that are used to swap in and out of other DH-200 and DH-220 type amps. Each time they improve the heft and solidity of the sound. Supposedly some significant advances in cap technology have been made in the last few years. Big old caps may sound flabby. And, they all do age. How old are the big old caps you now have? The newer caps can get more uF in a smaller space and do so with low ESR and DA (dielectric absorption). Big old caps are really cheap on eBay -- for a reason.

There no mods or small cap add-ons you will need to make for a MC PCB circuit. Don't do it. Also, note Hillig does use a 2 uF film cap across each PS cap. And, his PCB has a 1000 uF cap bypassing each PS rail and his circuit includes small film bypasses for most electrolytics. The PCB and power supply are in good shape with Hillig's suggested components. I would not deviate much from what he specifies.

Here is a picture of the rely and power resistor I use.

Cheers.........:angel:
 

Attachments

  • resistor-standoff.jpg
    resistor-standoff.jpg
    38.3 KB · Views: 164
Yes, that's me, Mr. Impatience. I also haven't been able to figure out yet how to make this forum quote a previous post, so I will create my own:

You wrote:
"I use one with a 5 second delay to bypass a 50 ohm 50 watt power resistor. That resistor, in series with the incoming AC for those few seconds let the big caps charge gracefully and no lights dim or switches blow up."

Thanks, I've heard about this but just understood the instruction better. Where do you position this delay in the case, and do you wire it right at the incoming AC line cord? Right now I am at 44K, so it may not be urgent to do this, but no doubt I will be looking for still more oomph at some point.

"Supposedly some significant advances in cap technology have been made in the last few years. Big old caps may sound flabby. And, they all do age. How old are the big old caps you now have?"

I don't really know, as I bought that unit already POOGED. Must be at least 10 years, could be 20. I also have a set of original Sangamos. They look fine, but they do hum some. So I am up for another set, in the 20-30K range.

"There no mods or small cap add-ons you will need to make for a MC PCB circuit. Don't do it."

I wasn't thinking about doing anything to the PCB circuits. I was talking about the big PS caps.

"Also, note Hillig does use a 2 uF film cap across each PS cap."

This is what I meant. The modification manual at http://www.musicaldesign.com/PA3C_1B.pdf does not show or mention such a cap that I could find. It is mentioned for the "Signature" version in the ad, and a photo of a modded unit on another page shows it. My AA mod uses a 5.1 uF Solen bypassed with a .47 across each big cap.

"And, his PCB has a 1000 uF cap bypassing each PS rail and his circuit includes small film bypasses for most electrolytics. The PCB and power supply are in good shape with Hillig's suggested components. I would not deviate much from what he specifies."

I wasn't thinking of deviating,...much. I am looking at something that I am not sure he has in his "basic" package, but is added at the next level. I may not know this for sure though, until I get it.

BTW, any reason I should not convert these Haflers to grounded cords when I change the cord?

Sorry about the impatience. My mother must have had a heck of a time holding me back when I was born.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.