I am looking to replace 2 CERHF-337 Capacitors in my Rockford Fosgate 200M Amplifier.
The capacitor size is 63V 330UF. Is there a replacement that provides more efficiency? More Power? etc?
Also, I need to replace the diode 4003 SS-161, not sure what to use here, the BUZ-103 that was burnt out, and the gain knob, which im unsure of what it is, except for the designation being RV1 on the board.
The capacitor size is 63V 330UF. Is there a replacement that provides more efficiency? More Power? etc?
Also, I need to replace the diode 4003 SS-161, not sure what to use here, the BUZ-103 that was burnt out, and the gain knob, which im unsure of what it is, except for the designation being RV1 on the board.
(Moreso a completely new post than an edit, but anyway..)
I need to replace both 330uf 63v capacitors (right next to the B+), (2) 2200uf 35v capacitors, 1 buz103 fet, the gain pot, and (2) PK IN 4003 0205 diodes.
can anyone give me some recommendations on which parts to replace them with?
should i change out the other, smaller, 35v caps with 105 degree ones?
i would also like to upgrade this amp while im making these replacements to learn more about amp repair, as i have about 10 more amps that i would like to get working again.
I need to replace both 330uf 63v capacitors (right next to the B+), (2) 2200uf 35v capacitors, 1 buz103 fet, the gain pot, and (2) PK IN 4003 0205 diodes.
can anyone give me some recommendations on which parts to replace them with?
should i change out the other, smaller, 35v caps with 105 degree ones?
i would also like to upgrade this amp while im making these replacements to learn more about amp repair, as i have about 10 more amps that i would like to get working again.
Electrolytic capacitors come in the 6 month life at room temperature variety, up to the 10 years in an oven motor compartment variety. I recommend the latter, unless you like doing work over. My DTV converter from W***-M*** came with bad electrolytics the day I bought it- always saving at W***-M***. The secret is buy caps rated at 105 deg C and 2000 hours up. 85 deg C caps can be good enough if the hours rating is good. Hours rating is correlated to the quality of the rubber used to seal in the water. Newark.com in NC has the hours rating in their selector table- click passive components, then capacitors, then aluminum electrolytic, then the value. If original was not +-20% or better tolerance, you can go up to 40% higher on size and 50% higher on voltage to get a bargain. Watch the size to see that it matches, also the radial-axial-snap in lead situation. If you are close to TX, mouser.com has significant cap stocks but you have to download the manufacturers datasheet to find out the cap life.
If I like a component, I usually do all the electrolytic capacitors at 20 to 25 years. These have a plus on one end or a minus on the other lead. Some say "NP" for non-polar. Match the tolerance (+-10% etc) in filter circuits. Mark the PCB with a sharpie which end is plus before you take them out -backwards blows up. After the amp sort of works, replace one component at a time between sounds tests to make sure you are making it better instead of worse. If worse, back up, think a while, ask questions here- you know which component made it worse, you just put it in. In values 10 uf or less, long life film caps have replaced electrolytic in the same size, and if size is a problem, I have used ceramic caps before for input caps although John Curl thinks this is stupid. Maximizing the voltage on ceramic caps minimizes the voltage driven non-linearity. I've been using Aerovox gold ceramic 50V caps for interstage coupling, which might have less quality problems than other brands since they are so expensive
Watch tantalum electrolytics, these usually have to be replaced by a film or aluminum cap in parallel with a .1 ceramic cap because of the speed. I don't buy tantalums, 100% if the ones I have bought have been defective, although John Curl that worked for an amp company had much better results at incoming inspection.
As far as sound goes, paralleling your main filter cap (probably the 330) with a .1 uf ceramic at the same voltage can add sparkle to your high frequencies in amps that don't have that already.
For the diode, if it is physically small (pea or smaller) in transistor devices I usually use a 400V rated fast recovery diode like the MR856 3 amp one. If it is right near a capacitor, it might be a zener diode, which has to be replaced by one of the same voltage and same or higher watt rating. Look up the part number on datasheetcatalog.com to find out what it really is- although in-house number won't be listed.
If I like a component, I usually do all the electrolytic capacitors at 20 to 25 years. These have a plus on one end or a minus on the other lead. Some say "NP" for non-polar. Match the tolerance (+-10% etc) in filter circuits. Mark the PCB with a sharpie which end is plus before you take them out -backwards blows up. After the amp sort of works, replace one component at a time between sounds tests to make sure you are making it better instead of worse. If worse, back up, think a while, ask questions here- you know which component made it worse, you just put it in. In values 10 uf or less, long life film caps have replaced electrolytic in the same size, and if size is a problem, I have used ceramic caps before for input caps although John Curl thinks this is stupid. Maximizing the voltage on ceramic caps minimizes the voltage driven non-linearity. I've been using Aerovox gold ceramic 50V caps for interstage coupling, which might have less quality problems than other brands since they are so expensive
Watch tantalum electrolytics, these usually have to be replaced by a film or aluminum cap in parallel with a .1 ceramic cap because of the speed. I don't buy tantalums, 100% if the ones I have bought have been defective, although John Curl that worked for an amp company had much better results at incoming inspection.
As far as sound goes, paralleling your main filter cap (probably the 330) with a .1 uf ceramic at the same voltage can add sparkle to your high frequencies in amps that don't have that already.
For the diode, if it is physically small (pea or smaller) in transistor devices I usually use a 400V rated fast recovery diode like the MR856 3 amp one. If it is right near a capacitor, it might be a zener diode, which has to be replaced by one of the same voltage and same or higher watt rating. Look up the part number on datasheetcatalog.com to find out what it really is- although in-house number won't be listed.
Last edited:
Okay, your second post came up while I was writing the first. 1N4003 is a garden variety 200V 1A rectifier diode, capable of being replaced by 1n4004-7 or any other diode with higher voltage and amp capacity. With BR856 running 4 for a dollar and shipping being $6.50 minimum, I keep a dozen BR856 (or equivalent, depending on what is cheap that day) around. Considering the price of Rockford Fosgate stuff, I doubt if your original electrolytic caps were of the 3000 hour design life variety.
On the pot, measure the shaft diameter and consider the lead mounting. Then pick a value. If you can't find a schematic, try 10k or 100k. If it is the volume knob, use audio or log taper, not linear.
Your fet, I don't know, but most FETs are E type N channel these days. If you try to match the package size (TO220 or TO3P or TO247 or TO3 metal case) the wattage might be close. Buy fets rated for higher voltage that your highest capacitor voltage, but not a whole lot higher (like don't use 1000 v fets in a car amp, they won't pass enough current). The higher the current rating the more you pay, and the safer your guess is if you can't find a datasheet on the originals (datasheetcatalog.com). If you have more then one output FET, then you need to replace them all for gain matching, I suppose.
On the pot, measure the shaft diameter and consider the lead mounting. Then pick a value. If you can't find a schematic, try 10k or 100k. If it is the volume knob, use audio or log taper, not linear.
Your fet, I don't know, but most FETs are E type N channel these days. If you try to match the package size (TO220 or TO3P or TO247 or TO3 metal case) the wattage might be close. Buy fets rated for higher voltage that your highest capacitor voltage, but not a whole lot higher (like don't use 1000 v fets in a car amp, they won't pass enough current). The higher the current rating the more you pay, and the safer your guess is if you can't find a datasheet on the originals (datasheetcatalog.com). If you have more then one output FET, then you need to replace them all for gain matching, I suppose.
Last edited:
I tried to replace a 1n4003 last night with a br856. The leads are too fat to fit in the PCB hole. So buy a fast recovery rectifier that is 1 or 2 amp to replace your 1n4003. Had a 95 VDC mistake blow through a 1n4003 backwards and blow 20 or so other semiconductors on a pcb as well as a cap and a resistor, so the 200V rating on a 1 cent part like a 1n4003 might not be the truth. Spend 30 cents on a 400V fast recovery rectifier from a name manufacturer like On, Fairchild, Vishay, ST or somebody.
On your FET, if it is near a toroid transformer by itself, it might be a switcher FET for making the 36VDC rails or something. Buy a 400V swticher FET for that, available from newark or mouser. If it is connected to the output in parallel with a lot of other FET's, it might be a linear output FET, so buy a FET whose datasheet mentions amplifier use whose transfer curve is more gradual. I haven't found those in stock yet, but cruise the solid state amp board for linear fet reccommendations. A lot of people recommend toshiba and sankyo and hitachi clones, and say the IR versions are just barely acceptable. But I am not a FEt expert yet. Tried to make a linear low power amp out of a switching type Fairchild FNC fet in place of a EF86 tube (600 V rating, right?) and the on point is so peaky that I couldn't get the DC bias circuit to have it act like an amplifier over varying temperature of warming up. Oh, buying fets, get the heat sink pad and thermal grease on the same order to save shipping charge. You might save the old pad, it might break, and a package of 5 of them is only $1.
On your FET, if it is near a toroid transformer by itself, it might be a switcher FET for making the 36VDC rails or something. Buy a 400V swticher FET for that, available from newark or mouser. If it is connected to the output in parallel with a lot of other FET's, it might be a linear output FET, so buy a FET whose datasheet mentions amplifier use whose transfer curve is more gradual. I haven't found those in stock yet, but cruise the solid state amp board for linear fet reccommendations. A lot of people recommend toshiba and sankyo and hitachi clones, and say the IR versions are just barely acceptable. But I am not a FEt expert yet. Tried to make a linear low power amp out of a switching type Fairchild FNC fet in place of a EF86 tube (600 V rating, right?) and the on point is so peaky that I couldn't get the DC bias circuit to have it act like an amplifier over varying temperature of warming up. Oh, buying fets, get the heat sink pad and thermal grease on the same order to save shipping charge. You might save the old pad, it might break, and a package of 5 of them is only $1.
Last edited:
sorry to hear about what happened with your diode. that sucks.
i've been speaking with matt roberts, and he has led to me some options for component replacements. please let me know what you think.
parallel 2 of these for each 330uf cap
Digi-Key - P10347-ND (Manufacturer - EEU-FC1J221S)
parallel 2 of these for each 2200uf cap
Digi-Key - P12406-ND (Manufacturer - EEU-FM1V122L)
diode replacement - but you said i should i think about going up to the 400v
Digi-Key - 1N4003FSCT-ND (Manufacturer - 1N4003)
then replace the buz103 mosfets with:
Digi-Key - FQP50N06-ND (Manufacturer - FQP50N06)
which lead me to wonder about running caps with a larger capacitance in parallel for each of the 330 and 2200uf caps, ie:
Digi-Key - P11249-ND (Manufacturer - EEU-FC1V392)
in the middle of our discussion he mentioned something about running some resistors in series to further lower noise/ripple. how would i go about determining which and how many resistors to place in front of the 2200uf caps(or their replacements)?
it's not that i'm stuck on the idea of increasing the capacitance, i just want to know if it would be better/worse/no noticeable difference. i'm assuming there is no sonic difference, but is there one in operational reliability or efficiency?
could i do something like getting one of those blank boards from radio shack and implement the resistor bridge on that and run wires from that to the pcb? i assume the largest issue with that would be space requirements and mounting issues.
would running larger capacitance in parallel with the resistor bridge be worth the time and effort?
im really enjoying learning about this stuff. as a kid, i took apart ever electronic device i could find. now im taking it a step further and improving upon them after i take them apart. or as the case is on the other 20 broken amps i have lying around here, just to fix em, get em out the door, and on to the next couple of broken amps.
i've been speaking with matt roberts, and he has led to me some options for component replacements. please let me know what you think.
parallel 2 of these for each 330uf cap
Digi-Key - P10347-ND (Manufacturer - EEU-FC1J221S)
parallel 2 of these for each 2200uf cap
Digi-Key - P12406-ND (Manufacturer - EEU-FM1V122L)
diode replacement - but you said i should i think about going up to the 400v
Digi-Key - 1N4003FSCT-ND (Manufacturer - 1N4003)
then replace the buz103 mosfets with:
Digi-Key - FQP50N06-ND (Manufacturer - FQP50N06)
which lead me to wonder about running caps with a larger capacitance in parallel for each of the 330 and 2200uf caps, ie:
Digi-Key - P11249-ND (Manufacturer - EEU-FC1V392)
in the middle of our discussion he mentioned something about running some resistors in series to further lower noise/ripple. how would i go about determining which and how many resistors to place in front of the 2200uf caps(or their replacements)?
it's not that i'm stuck on the idea of increasing the capacitance, i just want to know if it would be better/worse/no noticeable difference. i'm assuming there is no sonic difference, but is there one in operational reliability or efficiency?
could i do something like getting one of those blank boards from radio shack and implement the resistor bridge on that and run wires from that to the pcb? i assume the largest issue with that would be space requirements and mounting issues.
would running larger capacitance in parallel with the resistor bridge be worth the time and effort?
im really enjoying learning about this stuff. as a kid, i took apart ever electronic device i could find. now im taking it a step further and improving upon them after i take them apart. or as the case is on the other 20 broken amps i have lying around here, just to fix em, get em out the door, and on to the next couple of broken amps.
Read the download of that article. It does not quote the voltage rating of the ceramic caps related to the signal they pass. I'm running 50V ceramics on a 2 V pp signal, and find the results much more pleasing then the noisy tantalums caps I bought and used when I first repaired the amp in 1985. In fact the amp with the ceramic input caps (st120 with djoffe O.T bias circuit & modern fast Ft transistors) sounds very similar to another amp that has no caps in the signal path at all(CS800s). I have a pile of replaced tantalum caps in the back of the 1968 organ dated 2, 3, and 5 years after the organ was built. Presumably they were installed by dealer service. I'm guess production engineers are primarily interested in getting the life of the product past the warranty, but I'm more interested in updating stuff and leaving it alone for twenty years while I go on to something else. No more tantalum caps for me. And film caps are 50x the size of the tantalums or aluminum electrolytics I replace. Sometimes film doesn't fit.Do not use ceramic caps for audio coupling capacitors! We've all known it since Walt Jung and Richard Marsh 's landmark Audio magazine article in the 1980's (and letters by John Curl and Richard Marsh in the Audio Amatuer in the late 70's).
Downloaded the Panasonic EEU datasheet, which says they are 7000 hour caps. That sounds suitable. You should download the datasheet on every electrolytic you buy to ensure it is 2000 hours up. The 1N4003 is exactly the same as what blew up, so unless your shorted cap took it out, a higher voltage rating or current rating or a more reputable manufacturer's part should be used. Fairchild, the digikey link, is reputable, but a 1n4004 is a 400 V part. For more amps you should use the selector tables.
No opinion on the resistance trick in addition to bigger capacitors, except that if you increase load capacitance you have to do something about turn on surge not blowing up the input devices like rectifiers or fuses. I'm working on an amp now that has a turnon surge bad enough to make all the lights on the house circuit blink. Its breaker is tough enough to take it without tripping, but is also tough enough to not blow when the two DC detection crowbars melted the PCB lands to the outputs, and 15 of twenty Output transistors shorted or opened. I've supplimented the input breaker on this amp with fast blow fuses on the supply rails to limit the damage next time an output transistor produces DC on the output. Similar design updates should be done on your amp if you increase the capacitance of the power supply more than 40%.
Last edited:
Ceramic capacitors vary according to the composition of the dielectric, amonst other factors. They do seem to get a good kicking from some parts of the audio fraternity, yet do have their uses and like all capacitor types have their good points and bad points.
Also things have moved forward in material science and cap composition since the 1980's.
Also things have moved forward in material science and cap composition since the 1980's.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.