My Peavey 1.3k that I have been monkeying with for several months, has a shorted B side power transformer. Reads continuous on ohms scale, but collapses to 3 volts on one side of center tap if both B top and bottom are loaded with 700 ohms to center tap. I had first replaced the output caps, thinking maybe one was leaking (they are 18 years old)
I think I'm going to have to buy a new transformer. The original peaks at 95 VDC per rail, which I calculate at about 66 Vaverage (as transformers are sold). I'm actually reading above 90 VDC on the A side. My first thought was connect both output stages to the A side of the transformer and make a Peavey 650, but I'm afraid the A side would go assymetric with any load with a shorted turn in there and blow output transistors with the DC offset. There is only a 4k resistor pair establishing speaker return voltage. Any thoughts on that?
So the rated output is 51 V 8 ohm. I was wondering if modern transistors can run closer to the rail voltage without distorting much. I've actually installed MJ21195G and MJ21196G transistors, they are cheaper. Original OTs are MJ15024 and MJ15025 by the cross ref list. So maybe I could buy a lower voltage transformer or SMPS and waste less heat. 58 V rating instead of 68V?
I looked at the Peavey PV-2000 schematic for guidance, and it only has 3 O.T. pairs instead of 5, same part number!!! ???? The rail voltage is not listed on the PV-2000 schematic, which would be interesting to know.
I'm looking at Antekinc 1500 VA toroids and connexelectronic A1000SMPS with the stranded wire (1200 W) option. Various voltages are available. Any thoughts about the switching noise of the connexelectronic unit? Do I have to build a steel box around it? I've got plenty of junk finned heat sinks from blown motor drives.
(Oh BTW, I'm putting speaker disconnect relays in there based on the 8v dc offset diac triggering them. If I can fit it in I may put in relays between the rail capacitors and the O.T.'s to save money on OTs. My speakers cost 3X the cost of this beast with the parts, I wish to save them, not the amp.)
I think I'm going to have to buy a new transformer. The original peaks at 95 VDC per rail, which I calculate at about 66 Vaverage (as transformers are sold). I'm actually reading above 90 VDC on the A side. My first thought was connect both output stages to the A side of the transformer and make a Peavey 650, but I'm afraid the A side would go assymetric with any load with a shorted turn in there and blow output transistors with the DC offset. There is only a 4k resistor pair establishing speaker return voltage. Any thoughts on that?
So the rated output is 51 V 8 ohm. I was wondering if modern transistors can run closer to the rail voltage without distorting much. I've actually installed MJ21195G and MJ21196G transistors, they are cheaper. Original OTs are MJ15024 and MJ15025 by the cross ref list. So maybe I could buy a lower voltage transformer or SMPS and waste less heat. 58 V rating instead of 68V?
I looked at the Peavey PV-2000 schematic for guidance, and it only has 3 O.T. pairs instead of 5, same part number!!! ???? The rail voltage is not listed on the PV-2000 schematic, which would be interesting to know.
I'm looking at Antekinc 1500 VA toroids and connexelectronic A1000SMPS with the stranded wire (1200 W) option. Various voltages are available. Any thoughts about the switching noise of the connexelectronic unit? Do I have to build a steel box around it? I've got plenty of junk finned heat sinks from blown motor drives.
(Oh BTW, I'm putting speaker disconnect relays in there based on the 8v dc offset diac triggering them. If I can fit it in I may put in relays between the rail capacitors and the O.T.'s to save money on OTs. My speakers cost 3X the cost of this beast with the parts, I wish to save them, not the amp.)
Last edited:
Did you ask Peavey if they still have the transformer? Big honking thing like that has to fit and mount in the chassis. The factory part is going to bolt right in.
You cannot run both channels off one secondary winding. Remember this amp has "flying rails." The CT of the power transformer does not go to ground. The power supply is not just sitting there.
As to throwing a lot of transistors at it to get closer to the rails, I don;t think that is a good move. The amp ratings are not pushing things, any limitations on how close to rails you get are not due to a fraction of a volt junction potential in the transistors. It puts out its rated power.
I don't see 4k resistors in the speaker circuit. WHat are the resistor numbers you refer to and where are they on the print?
You cannot run both channels off one secondary winding. Remember this amp has "flying rails." The CT of the power transformer does not go to ground. The power supply is not just sitting there.
As to throwing a lot of transistors at it to get closer to the rails, I don;t think that is a good move. The amp ratings are not pushing things, any limitations on how close to rails you get are not due to a fraction of a volt junction potential in the transistors. It puts out its rated power.
I don't see 4k resistors in the speaker circuit. WHat are the resistor numbers you refer to and where are they on the print?
Thanks for listening. The center tap of the 95V on the PV1.3k is established by R265 and R267 on the diode bridge-rail capacitor PWB. They are 2 4k 5W resistors.Each winding CT connects directly to the emitters of the OT's on the two sides. Bottom of the eserviceinfo print.
Point taken about the flying rail whatever that means. I removed the A side capacitors and B side transformer leads, jumpered A side of PS to B side on the PS PWB, and tried to play it last night. The speaker on the B side was playing as well as the A side. The rail fuses (I installed the holders after getting tired of burned triac traces) were removed on the B side. ???? One datapoint is that the PNP collectors on the A side minus transistors are 650 ohms to the heat sink. Will be chasing that. Spent a couple of days removing rosin bridges to get the transistor quints to read <2ma current through a 47k resistor and a 19 V power supply from C to E on each quint with B to E shorted.
I don't hate my Peavey dealer but I've given up driving and he is 3 miles from the nearest bus stop down a lane with deep ditches, no sidewalks, and lots of impolite gravel & lumber trucks. I've put off walking up there to buy the double diode for the heat sink, and I really don't want to carry home a 40 lb transformer whatever it costs.
Besides, my speakers are either SP2-XT rated 49 Vrms, or T300 Hi-frequency-projectors rated not very much with 2 10" woofers. This 1.3k is really too hot to keep as is. If I sold it I'd only get $2-300. If I have to buy a transformer, I'll probably keep it for the T300's to play PBS great Performances on. So I want something a little less voltage spike inducing so I don't blow the cones out the grills. I really wanted a PV4 or something, but this came up first on craigslist. Drilling holes to mount weird transformers is not a problem. Actually I've already got 4 each @$10 132 VAC CT 2.5 A transformers that will fit in there. Some sort of 1 ohm balance resistors on each leg will be required to parallel 2 transformers, I suppose. Cut the turn on surge to 10000 uf, too.
SO why does the PV1.3k have 5 OT pairs per channel and the PV 2000 have 3? Noisier faster fan? Better heat sink design? Higher voltage rails?
Point taken about the flying rail whatever that means. I removed the A side capacitors and B side transformer leads, jumpered A side of PS to B side on the PS PWB, and tried to play it last night. The speaker on the B side was playing as well as the A side. The rail fuses (I installed the holders after getting tired of burned triac traces) were removed on the B side. ???? One datapoint is that the PNP collectors on the A side minus transistors are 650 ohms to the heat sink. Will be chasing that. Spent a couple of days removing rosin bridges to get the transistor quints to read <2ma current through a 47k resistor and a 19 V power supply from C to E on each quint with B to E shorted.
I don't hate my Peavey dealer but I've given up driving and he is 3 miles from the nearest bus stop down a lane with deep ditches, no sidewalks, and lots of impolite gravel & lumber trucks. I've put off walking up there to buy the double diode for the heat sink, and I really don't want to carry home a 40 lb transformer whatever it costs.
Besides, my speakers are either SP2-XT rated 49 Vrms, or T300 Hi-frequency-projectors rated not very much with 2 10" woofers. This 1.3k is really too hot to keep as is. If I sold it I'd only get $2-300. If I have to buy a transformer, I'll probably keep it for the T300's to play PBS great Performances on. So I want something a little less voltage spike inducing so I don't blow the cones out the grills. I really wanted a PV4 or something, but this came up first on craigslist. Drilling holes to mount weird transformers is not a problem. Actually I've already got 4 each @$10 132 VAC CT 2.5 A transformers that will fit in there. Some sort of 1 ohm balance resistors on each leg will be required to parallel 2 transformers, I suppose. Cut the turn on surge to 10000 uf, too.
SO why does the PV1.3k have 5 OT pairs per channel and the PV 2000 have 3? Noisier faster fan? Better heat sink design? Higher voltage rails?
Last edited:
No no no, pick up the phone and CALL Peavey in Mississippi. Then you only have to carry the part from your door, where UPS drops it off. Your dealer won;t have one, he'll have to call and order it anyway. PV sells direct to anyone. DOn;t walk anywhere with it other than in the house.
I think you will find those 4k resistors are just there to even things out, the transformer center tap is connected to the speaker lead at J103. They are not an artificial center tap - they are not like the pair of 100 ohm resistors across a Fender 6v winding.
FLying rails? Grounded emitter to some? Look at the schematic. Your common everyday amp has output transistors from the V+ and V- rails through emitter ballast resistors and meeting at the speaker hot lead. Now look at this one. That point is grounded. The "output" goes to ground. The speaker is wired from ground to the center tap of the power transformer. SO when signal is amplifying, the +/-95v rails don;t just sit there, they are moving around with the music. Having two channels trying to run off it might be done if they are getting the exact same signal, but if the signal differes between channels, then one channel tries to move it one way while the other tries to move the same supplies the other way. PFFFT. It might be a little more clear on the PV2000 drawing.
I don;t know which drawing eservice offers, I always use what the factory provides, in thos case drawing 98872002 for the PV1.3K.
Before you go trying to rewire this, you really ought to study how this system works. It is not just a power supply sitting there. You don;t want to burn up a new transformer.
I think you will find those 4k resistors are just there to even things out, the transformer center tap is connected to the speaker lead at J103. They are not an artificial center tap - they are not like the pair of 100 ohm resistors across a Fender 6v winding.
FLying rails? Grounded emitter to some? Look at the schematic. Your common everyday amp has output transistors from the V+ and V- rails through emitter ballast resistors and meeting at the speaker hot lead. Now look at this one. That point is grounded. The "output" goes to ground. The speaker is wired from ground to the center tap of the power transformer. SO when signal is amplifying, the +/-95v rails don;t just sit there, they are moving around with the music. Having two channels trying to run off it might be done if they are getting the exact same signal, but if the signal differes between channels, then one channel tries to move it one way while the other tries to move the same supplies the other way. PFFFT. It might be a little more clear on the PV2000 drawing.
I don;t know which drawing eservice offers, I always use what the factory provides, in thos case drawing 98872002 for the PV1.3K.
Before you go trying to rewire this, you really ought to study how this system works. It is not just a power supply sitting there. You don;t want to burn up a new transformer.
Okay, thanks. I didn't understand the flying rails. The schematic I've got doesn't show where the tabs between PCB's connect, so I wrote the colors down on the PCBs with a sharpie, but that doesn't organize the concept in my mind all that well. Actually, the schematic doesn't have the transformer PN either. Calling Peavey- that is a thought. I thought Peavey would be stuffy about selling to dealers only.
Abandoning the jumpered PS rails off one winding, and beginning install of two independent 132 VCT (open circuit) transformers. About 96 VCT at 2.5 A (tested). The HDTV output is real stereo, so independent windings will be required. Don't need 1.3 kw, 200 w/channel 8 ohms will be fine. Thanks again.
Abandoning the jumpered PS rails off one winding, and beginning install of two independent 132 VCT (open circuit) transformers. About 96 VCT at 2.5 A (tested). The HDTV output is real stereo, so independent windings will be required. Don't need 1.3 kw, 200 w/channel 8 ohms will be fine. Thanks again.
Last edited:
PV is absolutely THE most customer friendly comany in this business. They will sell you parts, they will send you schematics, you can call them on the phone for technical assistance, both for how to use their products and even in repair efforts.
I hope the experiment works.
If you can supply 2.5A into the load, why not calculate how much power that represents for say 8 ohms. Then calculate the current you need for the 200 watts. Remember, this is a solid state amp, it is going to produce an output signal voltage, then the load impedance will draw the current that demands... or at least try to. If these transformers can produce the full voltage the much larger stock transformer produces, then at full volume the load will try to draw the same power as well. The load has no idea the new transformer has a lot less current capability, it will still try to draw full current.
You still have the original transformer? The part number should be printed on it. PV transformer numbers start with 705.
I hope the experiment works.
If you can supply 2.5A into the load, why not calculate how much power that represents for say 8 ohms. Then calculate the current you need for the 200 watts. Remember, this is a solid state amp, it is going to produce an output signal voltage, then the load impedance will draw the current that demands... or at least try to. If these transformers can produce the full voltage the much larger stock transformer produces, then at full volume the load will try to draw the same power as well. The load has no idea the new transformer has a lot less current capability, it will still try to draw full current.
You still have the original transformer? The part number should be printed on it. PV transformer numbers start with 705.
I've blown up some stuff on the B driver board with my short A center to B center, drive A with radio & not B, no power to B OT's experiment. Never ask a question on Super Bowl Sunday and hope for an answer before Monday. I'm probably the the only man in the USA that doesn't watch pro football.
1.3k OT's and driver transistors are passing the 19V no leak current test(b-e shorted) but at least the I blew C213, and the OT Base driveline 5A fuse I installed instead of jumpers, and the protection 3.3 V 1.3 W zener diodes I put between the Emmitter drive traces and ground (to keep from melting ground traces on the driver PWB when an OT goes). Will be lifting a lot of diodes & stuff near the emmiter line on the B driver PCB looking for blown stuff for a few hours.
I don't even need 200 watts for my HDTV, just that 2.5A @ 95VAC these US made transformers put out is 250 watts and some of that goes in the heat sink. I'm quite happy with the volume I get out of my dynakit ST120 for home use, 60 W/ch. I'm running 1.5 Vpp at normal listening volume. into the SP2-XT's at 8 ohms. The T300 Projector speakers near the HDTV are 8 ohms too. In the PV 1.3k I had installed 4 ea 4700 uf 100v 2011 manuf caps on the B side PS, and 6 ea 3300 uf 100V 2008 manufactured caps on the A side power supply. I'll probably delete half the 4700's and two of the 3300s to unstiffen the power supply for home use. I might even delete 2 pairs of O.T. per side, I don't need that much current on the speakers. Actually I'm still a little worried about lightning pops causing the speakers to blow even with 2x 6600 uf rail caps. I get them through the ST120 even though I have a 600V MOS supressor installed on the ST120 transformer input in addition to the dynakit designed .01 uf 1000V disk cap. I've blown two unmatchable tweeters on 2 sets of speakers with a tube ST70 amp, so I might be hypersensitive to blowing stuff. At least the SP2-XT's have available tweeter drivers, which is part of why I bought them.
Even if I just use one 2.5A@95VCT transformer per side (Open circuit 132 VAC) I'm going to put 1 ohm 25 watt resistors on the t-former hot leads to cut the turn on surge and unstiffen the power supply. I'm hoping the DDT circuit will cut the impact of lighting pops on the speakers, but who knows how to test that it works. Everything around the 3180 IC's was blown on A channel including the J174 JFET, but the 3180's weren't shorted between pins like the 4558 op amp was.
I'll figure out all what is blown before I call Peavey for the double diode heat sink sensors. I don't have long distance telephone service; I make such calls from the payphone at gas station a mile away where it is hard to hear, so thanks for your help over the internet.
1.3k OT's and driver transistors are passing the 19V no leak current test(b-e shorted) but at least the I blew C213, and the OT Base driveline 5A fuse I installed instead of jumpers, and the protection 3.3 V 1.3 W zener diodes I put between the Emmitter drive traces and ground (to keep from melting ground traces on the driver PWB when an OT goes). Will be lifting a lot of diodes & stuff near the emmiter line on the B driver PCB looking for blown stuff for a few hours.
I don't even need 200 watts for my HDTV, just that 2.5A @ 95VAC these US made transformers put out is 250 watts and some of that goes in the heat sink. I'm quite happy with the volume I get out of my dynakit ST120 for home use, 60 W/ch. I'm running 1.5 Vpp at normal listening volume. into the SP2-XT's at 8 ohms. The T300 Projector speakers near the HDTV are 8 ohms too. In the PV 1.3k I had installed 4 ea 4700 uf 100v 2011 manuf caps on the B side PS, and 6 ea 3300 uf 100V 2008 manufactured caps on the A side power supply. I'll probably delete half the 4700's and two of the 3300s to unstiffen the power supply for home use. I might even delete 2 pairs of O.T. per side, I don't need that much current on the speakers. Actually I'm still a little worried about lightning pops causing the speakers to blow even with 2x 6600 uf rail caps. I get them through the ST120 even though I have a 600V MOS supressor installed on the ST120 transformer input in addition to the dynakit designed .01 uf 1000V disk cap. I've blown two unmatchable tweeters on 2 sets of speakers with a tube ST70 amp, so I might be hypersensitive to blowing stuff. At least the SP2-XT's have available tweeter drivers, which is part of why I bought them.
Even if I just use one 2.5A@95VCT transformer per side (Open circuit 132 VAC) I'm going to put 1 ohm 25 watt resistors on the t-former hot leads to cut the turn on surge and unstiffen the power supply. I'm hoping the DDT circuit will cut the impact of lighting pops on the speakers, but who knows how to test that it works. Everything around the 3180 IC's was blown on A channel including the J174 JFET, but the 3180's weren't shorted between pins like the 4558 op amp was.
I'll figure out all what is blown before I call Peavey for the double diode heat sink sensors. I don't have long distance telephone service; I make such calls from the payphone at gas station a mile away where it is hard to hear, so thanks for your help over the internet.
Last edited:
Okay, thanks cwahls for the new transformer. Works. the next challenge was the ground rivit was loose on B side O.T. assy so I had the rails going +176V -2 v instead of +-84 for a while, with the speaker return at -76V. (I do have the 120 vAC on a 1500 watt heater element which damps down the violence a little if an O.T. explodes).I drilled the rivit out and put a 4-40 stainless screw in there with star washers on each side. Rails are now +-84Vdc and speaker return is back to 0. I have 10 ohm 450 watt resistors on A and B speaker outputs, with a car radio speaker across 5 ohms of each. The car radio speakers are protected by 50 uf 600 VDC motor caps in series.
So I can't get B side to make music at all. A just plays right along with the radio input. On B driver PCB, I found the Jfet shorted across, and the 3080 op amp shorted from an input to one power rail, and the 4558 op amp measured a little funny between pin 6 and the negative power supply, so I replaced them all. The 33k and other resistors around there measure okay. All the electrolytic caps in the entire amp are new. I've got the O.T's idling on B side at O v DC out, and if I touch pin 5 of the 4558 with a charged up .1 cap it pops in the speaker. By playing an FM radio into it at 4 VAC, turning the input pot all the way up on B (pos 1 on A), I can get music to go through the pin 2 of the 6 pin connector from the input board, through pin 1 out of the 4558, and then it stops dead right after C206. I took the input board out today to look for shorted or blown diodes, resistors, or traces, and nothing is wrong there. the 2.2 NP cap is new too. (2.2 uf 50 V ceramic). there is no diagram of that on eserviceinfo so I had to draw it out, and make sure corrosion wasn't stopping my signal.
So I checked with my 200000 ohm/v simpson 266 on the 2.5 VAC scale, ,and I can see 3 VAC music come out of the input op amp (4558) on B side pin 1, go through c206 the 20 uf cap into the 1.2 k input to pin 6 of the 4558, and it stops dead at the input of the resistor. The resistor reads 1.2 k, the 61 k resistors reads right, the other 1.2k and 61k resistors from O.T. feedback pin 9 read right, the op amp just sits there with 0vDC on 5 and 6 and 0.8V dc coming out of 7. Op amp power supplies are fine. The 4558 is new, the 3080 is new, the J174 is new. pins 9 of the output transistor board is reading 0 (the feedback) and pin 10 is reading 0. The power to the op amp is reading +-15 per spec. ????? Am going to have to button it up and clear off the table to start checking the power supply of my scope why the sweep quit working, but probably I don't think I can chase this current mode problem with that either if it does ever work again . ($40 B&k 2120 scope, worked fine till I left in on for 3 hours once. will replace E-caps and maybe blown diodes etc. beam sits still).
So I can't get B side to make music at all. A just plays right along with the radio input. On B driver PCB, I found the Jfet shorted across, and the 3080 op amp shorted from an input to one power rail, and the 4558 op amp measured a little funny between pin 6 and the negative power supply, so I replaced them all. The 33k and other resistors around there measure okay. All the electrolytic caps in the entire amp are new. I've got the O.T's idling on B side at O v DC out, and if I touch pin 5 of the 4558 with a charged up .1 cap it pops in the speaker. By playing an FM radio into it at 4 VAC, turning the input pot all the way up on B (pos 1 on A), I can get music to go through the pin 2 of the 6 pin connector from the input board, through pin 1 out of the 4558, and then it stops dead right after C206. I took the input board out today to look for shorted or blown diodes, resistors, or traces, and nothing is wrong there. the 2.2 NP cap is new too. (2.2 uf 50 V ceramic). there is no diagram of that on eserviceinfo so I had to draw it out, and make sure corrosion wasn't stopping my signal.
So I checked with my 200000 ohm/v simpson 266 on the 2.5 VAC scale, ,and I can see 3 VAC music come out of the input op amp (4558) on B side pin 1, go through c206 the 20 uf cap into the 1.2 k input to pin 6 of the 4558, and it stops dead at the input of the resistor. The resistor reads 1.2 k, the 61 k resistors reads right, the other 1.2k and 61k resistors from O.T. feedback pin 9 read right, the op amp just sits there with 0vDC on 5 and 6 and 0.8V dc coming out of 7. Op amp power supplies are fine. The 4558 is new, the 3080 is new, the J174 is new. pins 9 of the output transistor board is reading 0 (the feedback) and pin 10 is reading 0. The power to the op amp is reading +-15 per spec. ????? Am going to have to button it up and clear off the table to start checking the power supply of my scope why the sweep quit working, but probably I don't think I can chase this current mode problem with that either if it does ever work again . ($40 B&k 2120 scope, worked fine till I left in on for 3 hours once. will replace E-caps and maybe blown diodes etc. beam sits still).
Last edited:
Great news! I love recycling things to those that can use it.
I still have one (presumably) good driver board here......
Have not tied into the 8.5C yet but keep posting on your progress, you may just get something in the mail.
Don't over think this, it's a simple amp, if you don't have schematics contact Peavey, they are free. When it's up and running, it a damn stable amp for what you are using it for.
When they blow up, they only blow up so far back, not often do they take opamps.
I still have one (presumably) good driver board here......
Have not tied into the 8.5C yet but keep posting on your progress, you may just get something in the mail.
Don't over think this, it's a simple amp, if you don't have schematics contact Peavey, they are free. When it's up and running, it a damn stable amp for what you are using it for.
When they blow up, they only blow up so far back, not often do they take opamps.
Thanks for listening. this one has exploded 4558 op amps while I watch, driven by the O.T.'s. Hint- when newark is selling On semi O.T.'s for 1/2 price, and they have a 200 mw rating, that is not a keypunch error. 10 seconds at 3 ma idle current at 90 v and kerpow!! Thought I had found a bargain, then thought I had some assembly errors (not). Hint 2, old O.T.'s that pass the double diode test are not necessarily good. Even one that didn't leak at 18V ceo exploded. So, 20 new O.T.'s at $3.80 apiece. To stop the exploding driver transistors op amps, and ground trace burning, I have put 3.9V zeners on the OT base drive lines on the PWB, and 5 amp fuses on those lines, on the driver on + and on the OT on minus. Also a 5 amp fuse on the middle, allegedly "earth" on the Pwb. I had burnt that a couple of places before fuses. No telling how the zeners sound, nobody on here had an opinion.
Wish I could edit after 30 minutes. After buttoning up, pulled B driver PWB and looked at C206 carefully again. The PWB trace was burnt off under the nice shiny new c206 cap lead on the + side. So no connection. Resoldered, plugged back in and - - - big hum from transformer, music on A not B, +22 on B rail, so something else is wrong. Power off. Back to DC analysis.
I've found 25A @ 600 V Allen Bradley contactors that will fit in the box by the fan, and plan to put one in to disconnect the speakers instead of the stupid triac PCB burners. Replaces the stupid bimetallic thermal disconnects. I'm using the Peavey protection diac (8v)(replaced) but changed the resistor from 47k to 20 k and the capacitor from 2 uf to 4 uf to try to get 5 ma out of it to drive triac type optoisolators. Two optoisolators back to back (with diodes for voltage protection) will trigger a triac the same way even if the input is either +85 or -85 DC. Bought some schottky 100V diodes to back up the opto inputs. The relays are 120 VAC coil. Unfortunately, the relays are NO only, so I have to leave them on all the time and come up with an inverter circuit. Found a form C relay but it won't fit in the space. I might still have to put a 24 VAC CT transformer in there to make inverting relay logic isolated from the sound power supplies. I still have to disassemble the relays and see if I can coat the copper points with silver plumbing solder. But this beast may get under control, eventually. Thinking about putting hall effect magnetic sensors at the zobel coil to use the jfet to damp really big current spikes, but don't really understand how the jfet works. Thought the jfet pinched off the signal to plus rail with the gate at first, but the 3080 op amp drives the resistor leg of the jfet, not the gate. The DDT defeat switch drives the jfet junction I found today, defeat is switch closed. ???
Wish I could edit after 30 minutes. After buttoning up, pulled B driver PWB and looked at C206 carefully again. The PWB trace was burnt off under the nice shiny new c206 cap lead on the + side. So no connection. Resoldered, plugged back in and - - - big hum from transformer, music on A not B, +22 on B rail, so something else is wrong. Power off. Back to DC analysis.
I've found 25A @ 600 V Allen Bradley contactors that will fit in the box by the fan, and plan to put one in to disconnect the speakers instead of the stupid triac PCB burners. Replaces the stupid bimetallic thermal disconnects. I'm using the Peavey protection diac (8v)(replaced) but changed the resistor from 47k to 20 k and the capacitor from 2 uf to 4 uf to try to get 5 ma out of it to drive triac type optoisolators. Two optoisolators back to back (with diodes for voltage protection) will trigger a triac the same way even if the input is either +85 or -85 DC. Bought some schottky 100V diodes to back up the opto inputs. The relays are 120 VAC coil. Unfortunately, the relays are NO only, so I have to leave them on all the time and come up with an inverter circuit. Found a form C relay but it won't fit in the space. I might still have to put a 24 VAC CT transformer in there to make inverting relay logic isolated from the sound power supplies. I still have to disassemble the relays and see if I can coat the copper points with silver plumbing solder. But this beast may get under control, eventually. Thinking about putting hall effect magnetic sensors at the zobel coil to use the jfet to damp really big current spikes, but don't really understand how the jfet works. Thought the jfet pinched off the signal to plus rail with the gate at first, but the 3080 op amp drives the resistor leg of the jfet, not the gate. The DDT defeat switch drives the jfet junction I found today, defeat is switch closed. ???
When I see "resoldered xxx and plugged it back in," I want to run from the room.
Are you not at least using a bulb limiter when first powering up after ANY change? Or better yet, a variac and ammeter on the mains. Just plugging it in is a great way to blow up good new parts you just installed. And possibly cause even further damage.
Any time you make a change you must do this. You cannot say "Oh, I just installed those a minute ago, they must be good. ANy time you power up and something fails, you must reset the "Oh that's OK" list. A circuit fault can blow up other parts faster than any fuse, breaker, electronic measure, or your fingers. Always check.
We have a bit of shop philosophy here, "Never think up reasons not to check something."
Not asking you to change, but whenever I read "OT" it means "output transformer". I have to stop and realize you must mean output transistor. Perhaps that is just me. Or maybe that I spend a lot of time with tube amp people.
I'd also like to highly recommend making the amp work as stock first, and only then start modifying it. If you throw all your re-engineering into the middle of the repair process, it only will serve to confuse the effort. Some zener, for example, may limit some transistor, but it can also wind up masking the cause of the problem you are trying to solve.
Look at the compressor circuit again. The 3080 doesn;t drive the JFET at all. On the print, "pin 1" of the JFET is ground, the bottom pin is connected to pin 3 of U100, the amp's input. The output of the 3080 is ALSO wired to that pin 3. The 3080 and the JFET are wired in parallel, the 3080 doesn't drive it. JFET Q101 is a power up mute. It is driven - its gate is controlled by - the voltage at C107 as charged by R113 from the +16v rails, which ALSO powers the 3080. That is all the JFET does. the 3080 handles all the DDT work alone. They can interfere with one another, of course, but they don;t work each other. The DDT switch controls the 3080.
Are you not at least using a bulb limiter when first powering up after ANY change? Or better yet, a variac and ammeter on the mains. Just plugging it in is a great way to blow up good new parts you just installed. And possibly cause even further damage.
Any time you make a change you must do this. You cannot say "Oh, I just installed those a minute ago, they must be good. ANy time you power up and something fails, you must reset the "Oh that's OK" list. A circuit fault can blow up other parts faster than any fuse, breaker, electronic measure, or your fingers. Always check.
We have a bit of shop philosophy here, "Never think up reasons not to check something."
Not asking you to change, but whenever I read "OT" it means "output transformer". I have to stop and realize you must mean output transistor. Perhaps that is just me. Or maybe that I spend a lot of time with tube amp people.
I'd also like to highly recommend making the amp work as stock first, and only then start modifying it. If you throw all your re-engineering into the middle of the repair process, it only will serve to confuse the effort. Some zener, for example, may limit some transistor, but it can also wind up masking the cause of the problem you are trying to solve.
Look at the compressor circuit again. The 3080 doesn;t drive the JFET at all. On the print, "pin 1" of the JFET is ground, the bottom pin is connected to pin 3 of U100, the amp's input. The output of the 3080 is ALSO wired to that pin 3. The 3080 and the JFET are wired in parallel, the 3080 doesn't drive it. JFET Q101 is a power up mute. It is driven - its gate is controlled by - the voltage at C107 as charged by R113 from the +16v rails, which ALSO powers the 3080. That is all the JFET does. the 3080 handles all the DDT work alone. They can interfere with one another, of course, but they don;t work each other. The DDT switch controls the 3080.
Cwahls, the transformer was nicely packed. The post office probably regrets their "it fits it ships" policy. Came on a Saturday afternoon, too, which was ultra cool for $12.
Thanks, Enzo. I'm using a 1500 watt room heater element in series with the 120 VAC amp, which cut the size of the O.T. explosions at installation. 100 watt bulbs are illegal now, and I never owned any for the reason they have been banned. With the amp dimming the lights at turn on, I was afraid to use any smaller bulb as a series limiter. The zener clamps behind fuses on the O.T. base driver line was self defense, I got tired of trying to figure out what trace the shorted O.T. burnt off the PWB this time. Was a few weeks before I figured out the 18V EB shorted O.T. leakage test found bad O.T.'s better than the double diode DVM check. Then I had to find out the newark NJW21194 O.T.'s with 200 mw power rating were rejects after the PNP original transistors stopped blowing up in pairs with them.
Thanks for the explanation on the JFET. I read in Mr Peavey's corporate history "white paper" that the DDT was invented to keep people from melting the speaker voice coils with square waves from their distorted guitar head. Nice, but I don't understand how these detect too much high frequency stuff (which is my understanding of a square wave). I don't see a RC filter for low frequencies. All I know at this point is the jfet and the 3080 have been blown on both sides, as I bought it (pretty destroyed). If I'm going to keep this thing, my worry is the lightning pop that comes at the beginning of a thunderstorm, not destroying the tweeters in my T300 HF projectors. Once I notice a storm is coming I turn stuff off (except tube amps) and this amp seems to be able to produce 1/2 watt with the input turned down, but all that energy is stored in there ready to come out at in inadvertant input. That is why I was musing about the zobel over current sense.
The different R & C are in there at the old diac circuit place for the relay disconnect, since I didn't want to take the O.T. assembly out again, and the triac traces were all melted off. The triacs are not ther so the diac and new RC are not doing anything to the circuit but driving a disconnected 2 wire cable at this point. When the amp works I'll look at putting the speaker disconnects relay drivers at the end of the cables. At this point (other than working on my failing internet connection today, and a bad hard drive) I'm going to replace the rectifier bridge on the B side that measures okay with the DVM, and doesn't leak current with the 18V supply. Something is collapsing the rail voltage on the B side when music happens, and with a different transformer in there, it is probably not the transformer. When I disconnect the B driver PWB from the O.T. assy, the rails go back to +-85, but when I put the 18V supply on the OT to driver collector inputs pin 3 and 6 (+-95 V nominal to driver) the driver is only leaking 0.3 ma. Thanks for listening.
Thanks, Enzo. I'm using a 1500 watt room heater element in series with the 120 VAC amp, which cut the size of the O.T. explosions at installation. 100 watt bulbs are illegal now, and I never owned any for the reason they have been banned. With the amp dimming the lights at turn on, I was afraid to use any smaller bulb as a series limiter. The zener clamps behind fuses on the O.T. base driver line was self defense, I got tired of trying to figure out what trace the shorted O.T. burnt off the PWB this time. Was a few weeks before I figured out the 18V EB shorted O.T. leakage test found bad O.T.'s better than the double diode DVM check. Then I had to find out the newark NJW21194 O.T.'s with 200 mw power rating were rejects after the PNP original transistors stopped blowing up in pairs with them.
Thanks for the explanation on the JFET. I read in Mr Peavey's corporate history "white paper" that the DDT was invented to keep people from melting the speaker voice coils with square waves from their distorted guitar head. Nice, but I don't understand how these detect too much high frequency stuff (which is my understanding of a square wave). I don't see a RC filter for low frequencies. All I know at this point is the jfet and the 3080 have been blown on both sides, as I bought it (pretty destroyed). If I'm going to keep this thing, my worry is the lightning pop that comes at the beginning of a thunderstorm, not destroying the tweeters in my T300 HF projectors. Once I notice a storm is coming I turn stuff off (except tube amps) and this amp seems to be able to produce 1/2 watt with the input turned down, but all that energy is stored in there ready to come out at in inadvertant input. That is why I was musing about the zobel over current sense.
The different R & C are in there at the old diac circuit place for the relay disconnect, since I didn't want to take the O.T. assembly out again, and the triac traces were all melted off. The triacs are not ther so the diac and new RC are not doing anything to the circuit but driving a disconnected 2 wire cable at this point. When the amp works I'll look at putting the speaker disconnects relay drivers at the end of the cables. At this point (other than working on my failing internet connection today, and a bad hard drive) I'm going to replace the rectifier bridge on the B side that measures okay with the DVM, and doesn't leak current with the 18V supply. Something is collapsing the rail voltage on the B side when music happens, and with a different transformer in there, it is probably not the transformer. When I disconnect the B driver PWB from the O.T. assy, the rails go back to +-85, but when I put the 18V supply on the OT to driver collector inputs pin 3 and 6 (+-95 V nominal to driver) the driver is only leaking 0.3 ma. Thanks for listening.
Last edited:
DO you have a variac and AC current meter?
The bulb thing works well enough, but if the amp is loo large for it, I have to wonder if putting some high wattage low resistance in series does the same thing. A variac and ammeter allows you to advance the mains to say 20v and see that current is already ramping up, so you turnit back off without ever stressing the stuff in the circuits.
Oh, and they phased out incandescent 100 watt bulbs in general, but decorative ones are still sold - you know the large round ones for vanity mirrors and such. And look at the dollar store.
The bulb thing works well enough, but if the amp is loo large for it, I have to wonder if putting some high wattage low resistance in series does the same thing. A variac and ammeter allows you to advance the mains to say 20v and see that current is already ramping up, so you turnit back off without ever stressing the stuff in the circuits.
Oh, and they phased out incandescent 100 watt bulbs in general, but decorative ones are still sold - you know the large round ones for vanity mirrors and such. And look at the dollar store.
Well, 100 W bulbs were back in the store today, now made in c****. Surprise. Now that I am not blowing up O.T.'s (output transistors) anymore. The room heater made a nice crackly noise when an O.T. blew, as the rails dropped to 40V or below. I got the room heater idea from Skippy/Dieter in Sydney, Aus. The last time I had B channel on music input with the resulting +3v -70V rails on the B side with 40VDC on the 10 ohm speaker simulator resistor, the heater was not crackling, and the 10 amp rail fuses (I installed) did not blow, so I am not blaming that on bad O.T.',s particularly as the +- rails went back where they belonged with the driver PWB unplugged from the OT assy. Have the generic replacement bridge out for installation but some PWB drilling required to wire in the + and - spade tabs. Had the 4560 op amp unplugged that time as I was under the impression it was just an LED driver for the front panel, but since + and - driver transistor input currents are inputs to the 4560, that was a stupid idea. Now that I traced the input board wiring this week. As the logic transistor on the driver PWB was also blown, and 3 driver transistors and both bias current clamp TO92 transistors were blown, plus a bunch of 50V capacitors and 1/4 w resistors, I'll have to suspect the 4560 is blown, too. I have some TI RC4560's. I think this band had a tendency to plug high power guitar amps into PA amp input phone jacks. As did the band that sold me the CS800s, also.
I realize, in full production, the rail fuses would have to be 25 amp, but 25 amp car fuses did not melt with shorted O.T.'s, so this whole fuse thing is a ****shoot. Pulling both rail voltages and ground center with a 3 pole relay on DC detection would be cool, and I have that many 25 A relays, but I can't make two of them fit in the box.
Time out for life. I have to drive out to north Clark county tomorrow and see what the tornado took away or dropped on my country property. It is south of Marysville between devastated Henryville and Chelsea, IN. Nothing to lose out there but a refrigerator stove and a freezer, a trailer with a new roof and rotten floors and walls, a couple of '98 model riding mowers with new motors and repaired decks, and a working Hammond H112 organ. Oh, a Peavey wedge monitor speaker that came with the rest of the bar band stuff. Hope my neighbors are okay. None of them had a basement or root cellar, as my place did once. Hope Cwalhs was not in the path, as this storm skipped across Illinois, also. However, I think I70 is a bit further north.
I realize, in full production, the rail fuses would have to be 25 amp, but 25 amp car fuses did not melt with shorted O.T.'s, so this whole fuse thing is a ****shoot. Pulling both rail voltages and ground center with a 3 pole relay on DC detection would be cool, and I have that many 25 A relays, but I can't make two of them fit in the box.
Time out for life. I have to drive out to north Clark county tomorrow and see what the tornado took away or dropped on my country property. It is south of Marysville between devastated Henryville and Chelsea, IN. Nothing to lose out there but a refrigerator stove and a freezer, a trailer with a new roof and rotten floors and walls, a couple of '98 model riding mowers with new motors and repaired decks, and a working Hammond H112 organ. Oh, a Peavey wedge monitor speaker that came with the rest of the bar band stuff. Hope my neighbors are okay. None of them had a basement or root cellar, as my place did once. Hope Cwalhs was not in the path, as this storm skipped across Illinois, also. However, I think I70 is a bit further north.
Last edited:
Well, everything is okay at my country property. The damage was 2 miles north of there. I saw Michigan had some problems, I see you (Enzo) are active on Solid State thread so I hope everything was okay with you- I think the tornado was in Ann Arbor.
Anyway, on the PV1.3k on your schematic, what do you have for the value of R102? My schematic says 4.7k but I have boards with professional looking 1.5k resistors installed for R102. I've been merrily blowing 4558 op amps on channel B and I'm getting 9 ma into 10 ohms out of U100 pin 7 without the IC, with the output transistor board disconnected, so it is coming out of R102, not the NPN predriver transistor. The NJM4558 datasheet shows output swing voltage degrading at 2000 ohms load, so I'm wondering if 1.5k is a little low resistance. If U100 was actually NJM4560 this much current wouldn't be a problem but they obviously got away from the 4560 for a reason. I suspect it is because 4560 might oscillate with that long run of parallel wires between input and output running over to the input board. No telling what IC was in this unit originally, the 4558's weren't NJM so probably were repairman installed. I got my "new" 4558's from mcmelectronics.com, so they should be "new" and the one on driver board A works okay with the 1.5k resistor. I'm getting 14V out 7 of this IC installed with 5 and 6 0V, which was new not too long ago. Trying 2.2k for R102 before I hear from you.
Anyway, on the PV1.3k on your schematic, what do you have for the value of R102? My schematic says 4.7k but I have boards with professional looking 1.5k resistors installed for R102. I've been merrily blowing 4558 op amps on channel B and I'm getting 9 ma into 10 ohms out of U100 pin 7 without the IC, with the output transistor board disconnected, so it is coming out of R102, not the NPN predriver transistor. The NJM4558 datasheet shows output swing voltage degrading at 2000 ohms load, so I'm wondering if 1.5k is a little low resistance. If U100 was actually NJM4560 this much current wouldn't be a problem but they obviously got away from the 4560 for a reason. I suspect it is because 4560 might oscillate with that long run of parallel wires between input and output running over to the input board. No telling what IC was in this unit originally, the 4558's weren't NJM so probably were repairman installed. I got my "new" 4558's from mcmelectronics.com, so they should be "new" and the one on driver board A works okay with the 1.5k resistor. I'm getting 14V out 7 of this IC installed with 5 and 6 0V, which was new not too long ago. Trying 2.2k for R102 before I hear from you.
Last edited:
R102, 1.5k or 4.7k? See the service bulletin I attached. Looks like your schematic is newer than your board, so make the changes to your board.
But I seriously doubt that change is the cause of your blow-ups.
If the schematic says 4558, then that is what belongs there. Especially since the U102 is specified as a 4560.
NJM doesn't matter, any 4558 will be fine. Leave the IC brand snobbery to the distortion pedal guys. These circuits are not that high-strung.
Let me get this straight, you are operating the amp with the power board disconnected? The 10 pin ribbon is off? DOn;t do that, the circuit depends upon that other board being there. You have removed some of the current paths, not to mention the dual diode between pins 1 and 2 that complete the bias string.
But I seriously doubt that change is the cause of your blow-ups.
If the schematic says 4558, then that is what belongs there. Especially since the U102 is specified as a 4560.
NJM doesn't matter, any 4558 will be fine. Leave the IC brand snobbery to the distortion pedal guys. These circuits are not that high-strung.
Let me get this straight, you are operating the amp with the power board disconnected? The 10 pin ribbon is off? DOn;t do that, the circuit depends upon that other board being there. You have removed some of the current paths, not to mention the dual diode between pins 1 and 2 that complete the bias string.
Thanks for service bulletin, thought the change might include changing 4558 to 4560 but I see not.
With the room heater in the 120 VAC line, I'm still getting collapse of rail voltages on B side driver board plugged in to +1.5V -80V. It is not blowing the 10 amp rail fuses, but the output transistors hold off the +-80V fine without the B driver board plugged in. I've already put stainless star washers on stainless screws between the speaker negative and the heat sink (ground) replaced the 10000 uf caps, the transformer and the diode bridge. No change. I'm trying to fix the driver board, So I try it with the output board unplugged for a second or two while I make measurements. With or without the output transistors board plugged in to the driver PCB, I get 14 VDC out of pin 7 of U100 on my second 4558 with 5 and 6 at 0 v. I've had 3 different transistors in the Q100 spot and none fails my 18vDC leakage test backwards. I get 8 ma out of pin 7 U100 without the IC or OT board link, which I thought was excessive. I have no 33 ohm resistors, and there is a stack of capacitors over R110 (10 ohms) simulating a 0.015 uf C120 that was blown, so there will have to be a parts order before I resume. I'm using TIP41C and TIP42C for Q100 & Q104 because they have the same safe operating area at 80V as MJ15032-33 at $.38 , but I guess I'll go ahead and buy some $3 MJE15032 and 15033. Thanks. Will be a while, I have to buy lawnmower tires and 23 acres of fuel this month.
With the room heater in the 120 VAC line, I'm still getting collapse of rail voltages on B side driver board plugged in to +1.5V -80V. It is not blowing the 10 amp rail fuses, but the output transistors hold off the +-80V fine without the B driver board plugged in. I've already put stainless star washers on stainless screws between the speaker negative and the heat sink (ground) replaced the 10000 uf caps, the transformer and the diode bridge. No change. I'm trying to fix the driver board, So I try it with the output board unplugged for a second or two while I make measurements. With or without the output transistors board plugged in to the driver PCB, I get 14 VDC out of pin 7 of U100 on my second 4558 with 5 and 6 at 0 v. I've had 3 different transistors in the Q100 spot and none fails my 18vDC leakage test backwards. I get 8 ma out of pin 7 U100 without the IC or OT board link, which I thought was excessive. I have no 33 ohm resistors, and there is a stack of capacitors over R110 (10 ohms) simulating a 0.015 uf C120 that was blown, so there will have to be a parts order before I resume. I'm using TIP41C and TIP42C for Q100 & Q104 because they have the same safe operating area at 80V as MJ15032-33 at $.38 , but I guess I'll go ahead and buy some $3 MJE15032 and 15033. Thanks. Will be a while, I have to buy lawnmower tires and 23 acres of fuel this month.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.