Looking for advice with this amp but here's the story 1st.
I just got this Hafler Pro4000 yesterday via Ebay and the guy says that he bought it new but never used it. I'm guessing that he bought it mid 2000's as the box has a 2004 date on it.
Anyway I hooked it up yesterday morning and used it all day and into the night powering a pair of JBL 2245s and it was fine.
This morning I turned it on and it was good for about 1 1/2 hours then I heard a loud pop from the left 2245 sounding like DC plus the speaker was doing a boner! I took it out of the rack and found 60 vDC on the left output, not good.
I took the cover off and checked the power supply voltages and dc after rectifier (with the B+ and B- to the board disconnected) and they appear to be okay. I have no B+ on the board so I suspect a short. I pulled the bias jumper off and got ~1500ma! Spec is 300ma. With the bias jumper off, I now have B+ on the board. It doesn't blow any fuses and I don't see any burned areas on the board.
Since it's been sitting so long is it possible a pwr supply cap has failed. How likely is it for one of the output devices to fail?
I'm going to go pull the module off so I can get at the solder side of the caps and check them for being shorted.
Thanks for any advice. Click on each half of the schematic and it'll show better.
Scott
I just got this Hafler Pro4000 yesterday via Ebay and the guy says that he bought it new but never used it. I'm guessing that he bought it mid 2000's as the box has a 2004 date on it.
Anyway I hooked it up yesterday morning and used it all day and into the night powering a pair of JBL 2245s and it was fine.
This morning I turned it on and it was good for about 1 1/2 hours then I heard a loud pop from the left 2245 sounding like DC plus the speaker was doing a boner! I took it out of the rack and found 60 vDC on the left output, not good.
I took the cover off and checked the power supply voltages and dc after rectifier (with the B+ and B- to the board disconnected) and they appear to be okay. I have no B+ on the board so I suspect a short. I pulled the bias jumper off and got ~1500ma! Spec is 300ma. With the bias jumper off, I now have B+ on the board. It doesn't blow any fuses and I don't see any burned areas on the board.
Since it's been sitting so long is it possible a pwr supply cap has failed. How likely is it for one of the output devices to fail?
I'm going to go pull the module off so I can get at the solder side of the caps and check them for being shorted.
Thanks for any advice. Click on each half of the schematic and it'll show better.
Scott
Little more info
I've actually got negative 60vdc on the output and not positive. I've got continuity between the negative output and the output device side of the bias jumper (with the jumper off) and I think this is pulling the B+ down.
I've actually got negative 60vdc on the output and not positive. I've got continuity between the negative output and the output device side of the bias jumper (with the jumper off) and I think this is pulling the B+ down.
I put my money on you being lied to. Most likely the outputs are dead, and were dead when it went into the box. I seriously doubt a capacitor failure, but it's easy to check for.
It's possible I'm sure but the seller is being sounding very concerned.
I pulled the module and the caps are not shorted so I'm guessing outputs.
I've actually got normal negative power supply voltage but no positive voltage at the board.
Looks like I'm looking for Q15, Q115 or Q4 as shorted. So if I find one shorted I need to replace a matched pair + and - right? Or do I need to replace all 6 mosfets on the channel?
The right side legs of all 3 N channel mosfets have continuity to ground but I don't know which one is the cause because their all linked. As far as matching is concerned, do all the N need to matched or do you match a N with a P?
I'm thinking I may need to replace all 6 dammit!🙁
I'm thinking I may need to replace all 6 dammit!🙁
You'd need to remove them to check them. Personally I'd replace all six and match them, but you'd have to buy several of each and match them yourself. Most do not go through the trouble (and the N and P will never match perfectly) and everything works just fine. That being the case, my advice would be to replace all three of the Ns with parts from the same batch. (If you order them at the same time from a reputable supplier, they should be from the same batch. Digikey, Newark, Arrow, Future, and Mouser come to mind.)
Just keep in mind that something made them fail and you may have other problems. At a minimum check the bias pot. Make sure it's not open or intermittent. I'd probably just replace it anyways. Those are cheap and it is old.
As for the seller. I don't know. There's good and bad on ebay. But, I always treat everything as broken and assume I'll have to fix it. If I do get something that works, its a pleasant surprise. I don't know the seller and shouldn't jump to conclusions, but I'm inclined to think it was bad when it went in the box.
Just keep in mind that something made them fail and you may have other problems. At a minimum check the bias pot. Make sure it's not open or intermittent. I'd probably just replace it anyways. Those are cheap and it is old.
As for the seller. I don't know. There's good and bad on ebay. But, I always treat everything as broken and assume I'll have to fix it. If I do get something that works, its a pleasant surprise. I don't know the seller and shouldn't jump to conclusions, but I'm inclined to think it was bad when it went in the box.
Thanks FoMoCo.
I think replacing all the mosfets on that channel is good advise and I'd rather buy them as a matched set. Digi Key only has the 2SK1058 for $9.48 but I found a matched set of the N and P for $49 that looks like a good deal here> Tech DIY Company Store.
What about the driver transistors? should I replace them too? I read in some blurb that you should if a mosfet fails.
Edit: The drivers are a no brainier as their available at the same place for $2.20 per matched pair.
I'm going to go check the pot and the bias on the other channel, lol!
Thanks again,
Scott
I think replacing all the mosfets on that channel is good advise and I'd rather buy them as a matched set. Digi Key only has the 2SK1058 for $9.48 but I found a matched set of the N and P for $49 that looks like a good deal here> Tech DIY Company Store.
What about the driver transistors? should I replace them too? I read in some blurb that you should if a mosfet fails.
Edit: The drivers are a no brainier as their available at the same place for $2.20 per matched pair.
I'm going to go check the pot and the bias on the other channel, lol!
Thanks again,
Scott
Last edited:
It sure can't hurt to change the drivers. At least check them. But, at that price I'd just replace them.
I'd be wary of a matched N and P pair. You might be getting BS'ed on that. N and P will never match very well. They may be matched for just Ids at a certain Vgs, or not matched at all. Surely transconductance and capacitance will not match well. You may wish to ask how they were matched and post the response here.
I'd be wary of a matched N and P pair. You might be getting BS'ed on that. N and P will never match very well. They may be matched for just Ids at a certain Vgs, or not matched at all. Surely transconductance and capacitance will not match well. You may wish to ask how they were matched and post the response here.
I ordered that mosfet set and drivers and will ask them how they match them and I post that when I hear. I'm going to unsolder those today one at a time and see if I can pin point the short but I don't want to touch the bias pot yet because I want to catch it if it was the reason for the failure.
Yesterday I pulled out channel 2, flipped it over and checked the bias. It was at 260ma, a little low, and set it to spec a 300ma. I let it run for several hours and all was okay.
The damn thing was sounding really good when it was working and this is really being a bummer. The seller isn't replying to my last emails but I have his phone number and I'm hoping that he will pay for the parts if I get this fixed.
Thanks again
Yesterday I pulled out channel 2, flipped it over and checked the bias. It was at 260ma, a little low, and set it to spec a 300ma. I let it run for several hours and all was okay.
The damn thing was sounding really good when it was working and this is really being a bummer. The seller isn't replying to my last emails but I have his phone number and I'm hoping that he will pay for the parts if I get this fixed.
Thanks again
Q15 is 100% shorted but I can't find anything else wrong so I guess I'm going to replace all 3 N channel mosfets and maybe the 3 Ps too but I haven't decided yet. I'm going to run it on the bench for a long time before I put in my system.
The seller is being very helpful and is going to pay for the parts.🙂
The seller is being very helpful and is going to pay for the parts.🙂
Erm ..you've bought a Crappy Amp and it subsequently died.
IMO Cut your losses before expending yet more $$ pointlessly.
IMO Cut your losses before expending yet more $$ pointlessly.
Maybe I was wrong about the seller. Check for any bad connections (or cold solder joints) that may have been intermittent and would allow it to be working and then be jarred loose and not working before powering up after the repair.
That's not a crappy amp. While I would not have designed an amplifier that way, it's not crappy. In my opinion that grounding the output devices idea is gimmicky, but doesn't make it a crappy amp.
I don't think I bought a crappy amp. I have 4 P3000s, 3 P1500s, 1 Pro2400 and 5 of the older TO3 Haflers and have had no problems with any of them except for the normal age related issues. I expected no issues with this one so this IS a big bummer.
The amp was new in box but the seller had sold it just prior to me buying it and that buyer had decided not to keep it. Prior to that guy buying it, it had been unopened. When the seller got it back, he checked the bias and esr of the power caps and ran it at 75% power for two days straight with no issues before re-listing it. Then I bought it and after about 12 hours of using it straight the 1st day and 1 1/2 hours the 2nd day, it failed.
Here's the sale
Hafler Transnova P4000 Used to Be New Read | eBay
The seller is one of a two man company building boutique guitar amps in New Jersey: Indigo Amplification
I talked to him on the phone Friday for over and hour and he was really helpful. He explained to me the whole history of this amp and how he bought it for a studio project that was never fulfilled and the story of the 1st sale. We went through the schematic together trying to come up things to check and he's a miffed as I am. Awesome dude🙂
I think I do see a few areas that could stand to be re-flowed and C35 is one and although the solder looks good on the underside of the board it looks pretty porous on top. C36 looks suspect too. There's also some minor surface corrosion in some areas so the board needs to be cleaned.
Do you think if the 47pf silver mica cap (C35) is a problem that it could have caused Q15 to fail? And also has anyone seen this type of mosfet just fail out of the blue? My other amps with this type haven't had a problem.
This is a list of resistors I've checked and they all test good:
R73, R159, R34, R1, R140, R139, R31, R32, R141 and the bias pot tests good.
I found the schematic is incorrect regarding C22 and C32. They are reversed as how they are shown.
I'm waiting for the mosfets before I go further.
Scott
I don't think C35 is the problem. Just check it with a meter. It should measure open for resistance. Most multimeters have a problem with values that low, so a capicitance check might be off. No need for concern. If it bothers you, replace it, we're only talking pennies for that part.
I'd replace the outputs, drivers, and the bias pots. (They can be intermittent and will kill the outputs if they open.) While you're in there you might as reflow any suspicious looking solder joints.
I still want to know how the N and P outputs were "matched." I just don't see this being true.
I'd replace the outputs, drivers, and the bias pots. (They can be intermittent and will kill the outputs if they open.) While you're in there you might as reflow any suspicious looking solder joints.
I still want to know how the N and P outputs were "matched." I just don't see this being true.
The piher 200 ohm trimmer is being difficult to find. I looked ay Mouser and Digi-Key with no luck. I'll email about the mosfet matching in a minute. I think they just match the P and N to themselves as you were saying before.
Thanks again
Thanks again
Replaced the 6 power mosfets, 4 driver transistors and set the bias to 300ma. Also reflowed a few other suspect solder points.
Couldn't find the correct bias pot so I left it as it seems to be fine and the bias adjusted okay.
I don't have a dummy load setup yet so I connected a old guitar speaker that I don't care about and I'm running a ~300hz into it at a level that is still ear friendly.🙂
I'm letting it go like this for a few days and hope for the best.
Tech-diy never answered my email as to how they bias the mosfets. I'm not going to keep bugging them about it.
Scott
Couldn't find the correct bias pot so I left it as it seems to be fine and the bias adjusted okay.
I don't have a dummy load setup yet so I connected a old guitar speaker that I don't care about and I'm running a ~300hz into it at a level that is still ear friendly.🙂
I'm letting it go like this for a few days and hope for the best.
Tech-diy never answered my email as to how they bias the mosfets. I'm not going to keep bugging them about it.
Scott
- Status
- Not open for further replies.