Hi.
The resistor R47 - 22R1 has failed in one of my channels. Would anyone know its wattage rating? I can only go of its size, which is 6mm long, 2.4mm wide on the ends, and 2mm wide in the centre.
Thanks
The resistor R47 - 22R1 has failed in one of my channels. Would anyone know its wattage rating? I can only go of its size, which is 6mm long, 2.4mm wide on the ends, and 2mm wide in the centre.
Thanks
Attachments
Looks like a quarter watt.
With vishay etc rating resistor wattage these days at a temperature high enough to burn the board material, I would buy metal film or carbon film resistor of at least 2 watt. Nothing shorter than the original, which might make it 3 watt or higher. It is the driver leg to speaker which carries significant current.
With vishay etc rating resistor wattage these days at a temperature high enough to burn the board material, I would buy metal film or carbon film resistor of at least 2 watt. Nothing shorter than the original, which might make it 3 watt or higher. It is the driver leg to speaker which carries significant current.
Last edited:
Hey, thanks, indianjo. 👍 I'll see what I can find and report back. The amp's around 27 years old, I guess resistors have come on a bit since then.
Looks like red-red-brown-gold-brown (right?). That should be 22R1 1% if my memory is correct.
Edit: Had a better look at the picture now (and I rethink 😀). Also 22R doesn't seem off for that position in the circuit and a power rating of 0.25-0.6W depending on resistor type should be OK as well.
Edit: Had a better look at the picture now (and I rethink 😀). Also 22R doesn't seem off for that position in the circuit and a power rating of 0.25-0.6W depending on resistor type should be OK as well.
Thanks.
Yep, red-red-brown-gold-brown. 22R 1% is what I thought. So I've replaced the resistor with a .22ohm 1%, hooked the power amp back up, and when I start turning the gain up the sound starts to distort?
I should add, the sound is fine at lower volumes, it's only when I raise it. Could the resistor be causing this issue, or?
Yep, red-red-brown-gold-brown. 22R 1% is what I thought. So I've replaced the resistor with a .22ohm 1%, hooked the power amp back up, and when I start turning the gain up the sound starts to distort?
I should add, the sound is fine at lower volumes, it's only when I raise it. Could the resistor be causing this issue, or?
Last edited:
R47 is clearly 22r1 whereas pr2 is 0r22 . Bryston is using R as the decimal point. R47 as 0.22 could prevent Q21 from driving Q28 adequately.
Usually when output stage parts blow, a lot of other parts blow because high voltage leaks out the base line of output transistors to previous parts that are not rated for that voltage or current. Suggest you 1 unpowered check all transistor with diode scale that b-e and c-e are .55-.70 volts forwards and high resistance backwards. Will not go all the way to 9999 because current leaks out through other parts. Also check resistance values same or lower than marked, lower because of current leakage. This can find bad parts, but does not prove them good.
2. Get alligator clip lead to connect negative of DVM to center of amp, say junction of PR2 and PR4. You cannot use speaker ground because of the speaker relay. Build a light bulb box with a 60 or 75 w tungsten light bulb between the hot of the power cord and the hot of an AC socket. Plug amp into the light bulb box and the box into the wall. Using only one hand, check DC voltages through the circuit, starting at the back. Transistors and diodes should drop 0.6 v forwards (or minus for PNP) be no less than 1v c-e (or minus for PNP) Determine voltages by subtracting, not by using 2 probes of meter to cross parts. >25 v from one hand to the other can stop your heart. Wear no jewelry on hands wrists or neck, 1v can burn your skin to charcoal through a ring. Wear safety glasses in case of parts explosions. Voltages can be checked against the other channel, or in general be <1/3 or 1/2 of the rail voltages, which will be lower than nominal because of the light bulb. If the light bulb is on, a sign that something is shorted.
Once light bulb is out and voltages look normal, check DC voltage on speaker + <+-0,2 v and current with no signal across emitter resistors PR2 PR4 etc. is about 25 ma.
Then play music through a junk speaker protected from DC by series back to back 2200 UF capacitors. At low volume the protection capacitors will cause some slight distortion.
Usually when output stage parts blow, a lot of other parts blow because high voltage leaks out the base line of output transistors to previous parts that are not rated for that voltage or current. Suggest you 1 unpowered check all transistor with diode scale that b-e and c-e are .55-.70 volts forwards and high resistance backwards. Will not go all the way to 9999 because current leaks out through other parts. Also check resistance values same or lower than marked, lower because of current leakage. This can find bad parts, but does not prove them good.
2. Get alligator clip lead to connect negative of DVM to center of amp, say junction of PR2 and PR4. You cannot use speaker ground because of the speaker relay. Build a light bulb box with a 60 or 75 w tungsten light bulb between the hot of the power cord and the hot of an AC socket. Plug amp into the light bulb box and the box into the wall. Using only one hand, check DC voltages through the circuit, starting at the back. Transistors and diodes should drop 0.6 v forwards (or minus for PNP) be no less than 1v c-e (or minus for PNP) Determine voltages by subtracting, not by using 2 probes of meter to cross parts. >25 v from one hand to the other can stop your heart. Wear no jewelry on hands wrists or neck, 1v can burn your skin to charcoal through a ring. Wear safety glasses in case of parts explosions. Voltages can be checked against the other channel, or in general be <1/3 or 1/2 of the rail voltages, which will be lower than nominal because of the light bulb. If the light bulb is on, a sign that something is shorted.
Once light bulb is out and voltages look normal, check DC voltage on speaker + <+-0,2 v and current with no signal across emitter resistors PR2 PR4 etc. is about 25 ma.
Then play music through a junk speaker protected from DC by series back to back 2200 UF capacitors. At low volume the protection capacitors will cause some slight distortion.
Last edited:
Thanks again. From your initial advice, I placed an order for a, Vishay 22ohm 2W 1%, however this is on back order. So in the meantime I ordered, what I thought was a 22ohm 1/4W 1%, but accidentally ordered .22ohm. I only realised this was wrong after fixing it to the board. When I placed the order again for the correct ohm, I was sent .22ohm again from the supplier, who said this is what I needed. So I just wanted to check I wasn’t going mad.
In UK Farnell & RS are usually reliable suppliers. You pick the part off the internet, there is no phone contact. Do not order parts that are out of stock. They have a minimum, maybe L20-30 or they impose a handling charge. Suggest you do the DVM check of parts and order all bad ones at once to save freight. Output transistors are ordered in sets, not individually because different parts runs have different characteristics. Note on >20 year old amps, if used 2000 hours/year nearly all electrolytic capacitors are at or near end of life. You may not want to change them all at once particularly as amp is broken, I make about 10% bad solder joints and newbies may be worse. But after amp works properly change them 1 or 2 at a time then test to make sure you did not make a mistake. Mark the board with a sharpie for plus, if you put electrolytic caps in backwards they blow the vent and leak. Larger parts order gets you fewer freight charges. All new e-caps makes an amp run like a new one, possible exception potentiometers and connectors.
If you are going to do this more than once, I keep a small stock of MJE15030/31 MJE340/350, NJW1302/3281, MPSA06/56 around, also 1n4004 and 6 amp rectifiers & bridges.
If some common part like 22 ohm 2 watt is out of stock, and no alternatives available, I use a different supplier. I have the choice of farnell, digikey, mouser, alliedelec. I know digikey & mouser are on Europe but you are cut off from them by customs barrier. Do not let farnell ship across the ocean. they show those parts on a selection table but sometimes they take 8 weeks to get through customs.
If you are going to do this more than once, I keep a small stock of MJE15030/31 MJE340/350, NJW1302/3281, MPSA06/56 around, also 1n4004 and 6 amp rectifiers & bridges.
If some common part like 22 ohm 2 watt is out of stock, and no alternatives available, I use a different supplier. I have the choice of farnell, digikey, mouser, alliedelec. I know digikey & mouser are on Europe but you are cut off from them by customs barrier. Do not let farnell ship across the ocean. they show those parts on a selection table but sometimes they take 8 weeks to get through customs.
Last edited:
Thank you, that's all good advice. There are 2 channel's that have issues, once I resolve the first channel I may need advise on the second. This one fried the board around the MJE15031 PNP Transistor, so it could probably be beyond my pay grade repair capabilities. Yup, I had thought the cap's could be coming to the end of there life span. Even though I don't use the amp that much, I've had the old girl quite a few years now, a good 24 years, so I'll start getting some replacements together. Looks like there's 2x 63V 15000uf, 3x16V 470uf, 2x 35V 330uf, 2x50V 22uf per channel. I'm not an electronic engineer, just an electrician, but my soldering skills are quit good. Cheers
I have patched pcbs that have lands burnt in two. I used 28 gage solid wire, insulated, on the driver base, I would use 22 ga around the output transistors. I drill small holes through the board to stuff the wires in from component side then bend them flat over the surviving land. Then solder.
If the board material is burnt to carbon, this has to be cut out. Cannot be stopped from shorting.
If the board material is burnt to carbon, this has to be cut out. Cannot be stopped from shorting.
I do not know the thermal design of a Bryston 9b. If you decide to never run the amp at full power for more than a few minutes, perhaps you could risk an inferior part.
15000 uF is kind of an obsolete size. Manufacturers have started using the EIA preferred size series. Perhaps you would have an easier time finding a 14000 uf or 16000 uf capacitor in stock.
The number of pins is not important in the day of effective glues. Multiple pins were used to resist the shock of transport by bar bands that set up & tore down every gig. Now we can use silicon seal or wallboard adhesive to glue the capacitor down to the pcb.
15000 uF is kind of an obsolete size. Manufacturers have started using the EIA preferred size series. Perhaps you would have an easier time finding a 14000 uf or 16000 uf capacitor in stock.
The number of pins is not important in the day of effective glues. Multiple pins were used to resist the shock of transport by bar bands that set up & tore down every gig. Now we can use silicon seal or wallboard adhesive to glue the capacitor down to the pcb.
So, it not been as simple as I expected finding a cap to replace the, KMH -105 degrees C - 63V - 15000uF (21D06L 26). However, I have been told that the, 63MXG15000MEFCSN35X50 is a suitable replacement. I have another couple more inquires I am going to make before going ahead with this overhaul.
