1. Please take 30 seconds to register your free account to remove most ads, post topics, make friends, earn reward points at our store, and more!  
     
    TalkBass.com has been uniting the low end since 1998.  Join us! :)

Ampeg PF-50T Humbuster

Discussion in 'Amps and Cabs [BG]' started by mnats, Oct 2, 2017.


  1. mnats

    mnats Supporting Member

    Feb 2, 2013
    I have identified the source of line frequency hum in the Ampeg PF-50T bass amplifier and have found a way to reduce it by 20 dB using about $2 worth of electronic parts (plus approximately two hours of an experienced technician's time). With a stock PF-50T the line frequency level is 60 dB below the signal which squares with the specifications in the Owner's Manual. Here is the 'Transformer Balanced' (post power amplifier) line output with a 500 Hz 0.135 VRMS signal to the input jack:
    PF-50T_Flat_THD+N_small.
    Here's the same amplifier modified with the $2 in parts mentioned earlier using the same conditions as before:
    PF-50T__Modified_Flat_THD+N_small.
    Details are on my Ampeg PF-50T Humbuster page. Remember that dismantling - not to mention modifying - your amplifier will void the warranty. Tube amplifiers contain lethal voltages that remain even after being disconnected from power. Leave any work to an experienced electronics service technician.
     
    Haroldo, agedhorse, 12BitSlab and 4 others like this.
  2. High Camp

    High Camp

    Oct 3, 2013
    Netherlands
    minus 20 dB is quite a lot, impressive!
     
  3. Bertr

    Bertr

    May 6, 2013
    Brilliant mod!
    I have just modded mine as well, a bit differently: I did cut the rear leg of the diode, inserted a 3,3k between the back of that diode D29 and R128.
    From the back of the diode, a cap (220 uf) connects to ground.
    Between the 2 caps (the new and the original), there is now 3,3k+680 instead of 680.
    That way, I did not have to change the trimpot, which has still the usable range.
    Well done for finding the origin of that annoying hum!
     
    Haroldo and mnats like this.
  4. beans-on-toast

    beans-on-toast Supporting Member

    Aug 7, 2008
    The fix is well worth the $2 cost of admission.

    I would include your board revision number as a reference on your web page.
     
    Last edited: Oct 3, 2017
    Haroldo and 12BitSlab like this.
  5. AstroSonic

    AstroSonic Supporting Member

    Dec 10, 2009
    rural New Mexico
    A person has to wonder why the stock bias supply circuit was chosen. Does it have any advantages over other long known quieter designs?
     
  6. boristhespider7

    boristhespider7

    Jan 27, 2008
    UK
    Is the bias system on other Ampeg heads different eg. V4b?
     
  7. beans-on-toast

    beans-on-toast Supporting Member

    Aug 7, 2008
    No problem that degrades performance is acceptable to the end user.

    Why they used this particular circuit is something only the designer can answer. It’s a simple circuit, a modification improves it. It also adds to the cost. Inflexible cost constraints, down to the cent, drives designers nuts.


    There are many possible reasons for an issue. It can be a design error. Sometimes people make rookie mistakes. It happens. A circuit may look fine on paper and perform well in the lab. On the circuit board, problems can arise such as noise being inducted in another circuit. It can be also be caused by an out of spec component in the circuit. When any issue is identified, it gets addressed in a revision. This is quite common in the industry, not just Ampeg.

    It might be something that a newer revision has already addressed. Older amps are updated to the current revision if they come in for service. Public recalls tend to only be made if there’s a safety related issue.
     
    Last edited: Oct 4, 2017
    AstroSonic likes this.
  8. beans-on-toast

    beans-on-toast Supporting Member

    Aug 7, 2008
    No, it isn’t the same circuit as the one used in the V4B.
     
  9. BassmanPaul

    BassmanPaul Gold Supporting Member

    Aug 25, 2007
    Toronto Ontario Canada
    A lot of amplifier designs use half wave rectification for the bias supply. Some take the feed from one side of the high voltage winding, some from a dedicated bias winding. This leads to a ripple of 50Hz, for those on 50Hz mains or 60Hz for 60Hz mains.

    For my own designs I have used full wave rectification for decades. It’s very easy to implement.

    For a dedicated bias winding use an appropriately rated bridge rectifier - almost any. The result is a supply ripple of 100Hz or 120Hz which is easier to smooth out.

    For those derived from the ends of the HV winding, usually via a rectifier and a resistor, double up on the rectifier and resistor from the opposite end of the winding. Again the ripple become 100Hz or 120Hz and is more easily smoothed. Edit: Obviously the two feeds come together at the top of the first bias smoothing cap.

    I hope someone finds this useful. :)
     
    Last edited: Oct 5, 2017
    AstroSonic likes this.
  10. beans-on-toast

    beans-on-toast Supporting Member

    Aug 7, 2008
    Thanks Paul! It’s getting close to dinner and your explanation makes me think of feeding on smooth chocolate ripple ice cream in two possible sizes.

    FEECB8FB-5A67-43B0-9659-4A3516D78854.
     
    pcake, JimmyM and kobass like this.
  11. BassmanPaul

    BassmanPaul Gold Supporting Member

    Aug 25, 2007
    Toronto Ontario Canada

    Sod!! Doncha know I’m a damn Diabetic!!! :crying:
     
    JimmyM and beans-on-toast like this.
  12. mnats

    mnats Supporting Member

    Feb 2, 2013
    Thanks for your post! As we've discussed privately I will add your mod and photo to my page. It simulates well and requires much less work - and disassembly - than mine.
     
  13. JimmyM

    JimmyM

    Apr 11, 2005
    Apopka, FL
    Endorsing: Ampeg Amps, EMG Pickups
    I wish I could speak Schematic. I haven't really noticed any obvious out of the ordinary hum with mine, and in fact just enjoyed using it on a gig yesterday, but if this mod got Beans' attention, then it gets mine. And I would probably be down with trying one of these mods if I knew what the hell I was looking at or had even the foggiest clue what you guys are talking about.

    Do have some questions, though...

    1. We know that mnats' method #2 cuts the noise half as much as method #1. How much noise cut can you get from the Bertr method? Again, don't know much about what's going on here, but when I hear "Cut a leg and solder something to it," it sounds way more attractive to me than "Cut traces and drill holes."

    2. Is reduced line noise the only thing that happens? Does the tone get changed? Does it compromise any other functionality?

    3. Can this stuff be explained at paint-by-numbers level rather than engineer level? And please don't tell me you're already explaining it at paint-by-numbers level, otherwise I may have to book an emergency IQ test.

    :D
     
  14. Jefenator

    Jefenator

    Aug 22, 2008
    Oregon
    I'm just glad to hear these amps can be tinkered with. :thumbsup:
     
  15. beans-on-toast

    beans-on-toast Supporting Member

    Aug 7, 2008
    It occurred to me that there is a bias circuit board in the TB Portaflex Wiki that can be adapted for use in the PF-50T. Here is the wiki link: A Fixed Bias Circuit Board.

    The board was intended to be used to convert the earlier cathode biased B-15N amps to the later fixed bias revisions. It can be adapted for this application. The advantage of a piggyback circuit board is that it is safe and road worthy.

    Here's the circuit. The component values will have to be changed for the PF-50T. R36 will be replaced with a jumper wire. R35 will not be installed. Capacitors with the correct footprint will have to be found.


    Bias circuit.
    full bias schematic.

    This is the circuit board layout.
    full bias layout.


    This is the printed circuit board. The dimensions are 38mm square (1.5 inches). A suitable location to mount the board above the main PCB will have to be determined. This can be done with a cardboard cutout. Finding the space to retrofit a board can be difficult in some amps. There are four mounting points, plastic standoff can be used. If bus wire is used to run the connections to the main circuit board, the mounting is reinforced. The feet of the standoffs can be gooped to the board. To serve as an example, mounting a tone circuit board and the part numbers for the standoff hardware are here: Tone Circuit Replacement.

    full bias circuit V1.2 top.
     
  16. kobass

    kobass Supporting Member

    Nov 3, 2003
    Outside Boston
    Any amp can be tinkered with-the results may vary... :D
     
    agedhorse and mnats like this.
  17. AstroSonic

    AstroSonic Supporting Member

    Dec 10, 2009
    rural New Mexico
    Just maybe your amp incorporates changes to address this issue. There should be a board and revision number on the circuit board. It might be different than that on the boards that have this problem. If everybody could check and post their board numbers we might be able to figure this out.

    It seems to me that if a person is going to modify an amplifier, it should be done with a minimum of modification to the stock circuit board, and should be 'road ready'. The B15N after-market bias board would be acceptable if it fit securely and was available for purchase.
     
    JimmyM likes this.
  18. beans-on-toast

    beans-on-toast Supporting Member

    Aug 7, 2008
    If the noise level is normal in comparison to other amps, and the manufacturer says this is how it should be, it doesn’t make sense to modify it. If Loud has identified an issue and has a fix, that’s a different story.

    For those that want to tinker, why notif you can improve performance. I tend to modify all my personal amps. They all serve as testbeds.

    You can hear the noise in this amp when the bass is quiet, especially towards the end when the treble is boosted all the way. It is normal for some settings to bring out high frequency hiss. A tweeter magnifies it. It depends on the cab that you are using.

     
    Last edited: Oct 5, 2017
  19. mnats

    mnats Supporting Member

    Feb 2, 2013
    I think Bertr's fix (now on my Humbuster page) is the best solution as it uses the existing externally accessible trim pot and will still give the 20 dB reduction in noise according to simulation. I'm going to convert mine back to original so I can add that to the video on my page, then use Bertr's mod to do some measurements. There are a million ways to do a negative bias circuit; I don't think we need to know more ways to implement it for the purpose of fixing the hum in this model.

    About the hum, if you can't hear it and missed the warnings about cracking open your amplifier then you have two good reasons not to do it.

    Regarding the board revision number they are different on the solder and component sides for reasons unknown. I will get around to posting images but for now I think the best way to identify the problem is to listen to your PF-50T with the gain and volume all the way down. If you are going to open the amplifier to check PCB revision numbers you may just want to see if the small number of bias components that match my schematic are present. Component designations on my circuit are correct and correlate to my PCB.

    I also would like to point out that my amplifier as received matched the current Owner's Manual specification of a signal to noise ratio of 60 dB from 20 to 20kHz. Funny that @agedhorse was recently asking what specifications users wanted to see yet somehow this marginal specification gets overlooked...
     
    AstroSonic and JimmyM like this.
  20. JimmyM

    JimmyM

    Apr 11, 2005
    Apopka, FL
    Endorsing: Ampeg Amps, EMG Pickups
    Mine was made before it was released to the public so I doubt I'd be any help. Probably another reason I shouldn't tinker with it.