Power
From what I read, the Apt-1 provides 100wpc into 8 ohms and 200wpc into 4 ohms. There's an impedance matching switch on the back, one position for 2-4 ohms and the other for 8-16 ohms. I'm sure UncleJed can elaborate when he next checks in.
I think if you search back through this thread you'll find a link for an owners' manual. If not, lemme know and I can email you the pdf file.
From what I read, the Apt-1 provides 100wpc into 8 ohms and 200wpc into 4 ohms. There's an impedance matching switch on the back, one position for 2-4 ohms and the other for 8-16 ohms. I'm sure UncleJed can elaborate when he next checks in.
I think if you search back through this thread you'll find a link for an owners' manual. If not, lemme know and I can email you the pdf file.
APT-1 Further Problems 2
I reconnected the amp and found out that there is -60V at the output on the left channel. I also found out that there is the same -60 V on all three legs of the U60 pre and 2SA1112 driver. I took them out but they check ok. The 4007 diodes are ok.
The other drivers have - on E and B and + on C.
Any further suggestion?
What are the predrivers marked as U60 and U10? I have the schematics but not a list of the components.
I reconnected the amp and found out that there is -60V at the output on the left channel. I also found out that there is the same -60 V on all three legs of the U60 pre and 2SA1112 driver. I took them out but they check ok. The 4007 diodes are ok.
The other drivers have - on E and B and + on C.
Any further suggestion?
What are the predrivers marked as U60 and U10? I have the schematics but not a list of the components.
D'oh!
I tried that, and it seems to work like a charm! The impedence warning lamp only occasionally flickers on really loud passages, so all seems well there. And yes, it can now go louder than I care for it to. There are still some passages that sound strained, but that may well be the recording instead of the amp. I may well try the autoformers though and report back in.@Tekkir:
I do have the owner's manual, but thanks for the offer to upload it to me.
And though it's completely off topic, I'm using a Behringer DEQ2496 as a "room correction" unit with the Magnepan MMGs, and the synergy between the two is truly amazing. Just by using the EQ in subtractive mode (ie, reducing volume on the frequency bands vice increasing it) I can get the response in the room to +/- 3 dB from 40 Hz to 20K Hz! Not to mention the Maggie's trademark air, transient response and detail. Can't get this stupid grin off my face when listening.
And yeah, they do have slam for rock as well. My Apt gear (Apt-1 and Apt Holman pre-amp) go nicely in that system as well (not so off topic after all)Not shilling anything, just sharing my enthusiasm at discovering a match made in audio heaven. Back to listening...
MG-bert
Apt 1 op amp input stage modifications
Uncle Jed, whatever happened to the the op amp input stage mods for the apt-1? I have an apt-1 also, with an intermittent input connector thats waiting for a little work. I'd love to upgrade the input stage to get an extra 6db s/n. Let me know, apt-1's become available on ebay now and then at very reasonable prices. I also am driving Magnepan MMG speakers but with the switch in the 2-4 ohm position. Or was, currently using my backup, an adcom 535. Not quite in the same league.
Uncle Jed, whatever happened to the the op amp input stage mods for the apt-1? I have an apt-1 also, with an intermittent input connector thats waiting for a little work. I'd love to upgrade the input stage to get an extra 6db s/n. Let me know, apt-1's become available on ebay now and then at very reasonable prices. I also am driving Magnepan MMG speakers but with the switch in the 2-4 ohm position. Or was, currently using my backup, an adcom 535. Not quite in the same league.
ok, most pressing issue first...
NJ, if you have rail voltage at the output, and all the way back to the predrivers you may have a shorted VAS transistor on the - side or an open one on the + side or the current source for the diff amp failed. i need a listing of voltage measurements for all of the transistors. it would be wise to remove the output devices in that channel and replace a pair of them (one NPN and one PNP) with 1.5k 2 watt resistors across the B-E connections. this will give you a voltage output at the speaker terminal and keep the feedback loop intact without risking the lives of the output devices. at APT we had a pair of dead output transistors with the top hat cut off, the B-E bond wires cut and the resistor soldered to the B-E pins.
if i get questions like this enough i may make an Excel spread sheet....
the U10 and U60 transistors are MPSU10 and MPSU60 devices made by motorola and i think are no longer made, but ON should have something at least similar. their pinout is EBC unlike japanese devices which would be BCE.
as far as the op amp input stage goes, i don't recommend chopping your amp. i actually built the amp module from scratch and left out the components that were being bypassed. it would require a pair of blank amp boards and a kit of parts. not a job for the faint of heart... i suppose you could use the etch mask on the cover of the owner's manual, but that's for a REV 2 amp board.
NJ, if you have rail voltage at the output, and all the way back to the predrivers you may have a shorted VAS transistor on the - side or an open one on the + side or the current source for the diff amp failed. i need a listing of voltage measurements for all of the transistors. it would be wise to remove the output devices in that channel and replace a pair of them (one NPN and one PNP) with 1.5k 2 watt resistors across the B-E connections. this will give you a voltage output at the speaker terminal and keep the feedback loop intact without risking the lives of the output devices. at APT we had a pair of dead output transistors with the top hat cut off, the B-E bond wires cut and the resistor soldered to the B-E pins.
if i get questions like this enough i may make an Excel spread sheet....
the U10 and U60 transistors are MPSU10 and MPSU60 devices made by motorola and i think are no longer made, but ON should have something at least similar. their pinout is EBC unlike japanese devices which would be BCE.
as far as the op amp input stage goes, i don't recommend chopping your amp. i actually built the amp module from scratch and left out the components that were being bypassed. it would require a pair of blank amp boards and a kit of parts. not a job for the faint of heart... i suppose you could use the etch mask on the cover of the owner's manual, but that's for a REV 2 amp board.
Last edited:
I did look further at the board and Q2 and Q3 were burned out, as well as the two resistors between them. Other transistors seemed to be ok. I could replace the burned part but I am afraid that they could burn again since I haven't found the reason why they burned out. Any way to proceed safely?
APT-1 board mismarked? HELP!
As I continue to try to troubleshoot the APT-1 I have, I noticed a very puzzling thing! On my board, marked as Rev2, Q5 is identified as a D51 but a 2n3906 is soldered in its place (no big deal, both PNP...) also, Q6 is marked as a D01 (NPN) but a D51 (PNP) is in that place! This mismatch is probably the cause of all the trouble with burned out resistors and other transistors!
Does anybody have relevant info/advice about this problem? I look forward to a quick answer before further damage happens to the amp!
Antonio
As I continue to try to troubleshoot the APT-1 I have, I noticed a very puzzling thing! On my board, marked as Rev2, Q5 is identified as a D51 but a 2n3906 is soldered in its place (no big deal, both PNP...) also, Q6 is marked as a D01 (NPN) but a D51 (PNP) is in that place! This mismatch is probably the cause of all the trouble with burned out resistors and other transistors!
Does anybody have relevant info/advice about this problem? I look forward to a quick answer before further damage happens to the amp!
Antonio
heat sinks go on the D01 and D51 devices.... has somebody worked on the amp before and not got it working? to tell if it really is a rev2 board (some of the early rev 3's might have been reworked rev 2 boards) tell me if there is a capacitor near the D01 or D51 transistor standing on long leads (i remember that something on the top of the board may have been soldered to it, and that was the reason for the long leads). i no longer have my copy of the manual with all of the revs up to rev 10, but the revision changes on the boards were usually minor changes.
Modified Board?
The board I have looks modified in several points, although it is marked as Rev 2, while the input board is marked as rev 6. Indeed it had a long leads 47 uF, 35 V cap sticking out with a -12 orange wire soldered to one of the legs, I put it back into the board as I recapped. It also has a wire from the -75V fuse holder to a diode that is lifter from the board, some '37 zener diodes removed, some resistors mislabeled, etc. Because of the offset I had, i checked all transistors and replaced all the small ones ( drivers and predrivers tested ok) beside recapping. In replacing the transistors, I followed what written on the board (since the schematics I have are not so clear) and if indeed I have misplaced an NPN in place of a PNP, it would explain the burned out resistors an transistors I got. I realized this by comparing with the other channel that is luckily working...
Which 01 and 51 have the heat sink? there are 2 51 and 6 01 on the board....
Thank you for any advice you may have in fixing this old glory....
The board I have looks modified in several points, although it is marked as Rev 2, while the input board is marked as rev 6. Indeed it had a long leads 47 uF, 35 V cap sticking out with a -12 orange wire soldered to one of the legs, I put it back into the board as I recapped. It also has a wire from the -75V fuse holder to a diode that is lifter from the board, some '37 zener diodes removed, some resistors mislabeled, etc. Because of the offset I had, i checked all transistors and replaced all the small ones ( drivers and predrivers tested ok) beside recapping. In replacing the transistors, I followed what written on the board (since the schematics I have are not so clear) and if indeed I have misplaced an NPN in place of a PNP, it would explain the burned out resistors an transistors I got. I realized this by comparing with the other channel that is luckily working...
Which 01 and 51 have the heat sink? there are 2 51 and 6 01 on the board....
Thank you for any advice you may have in fixing this old glory....
also know that there may have been some errors in the artwork on the top of the board with the D01, D51, 3904 and 3906 transistors in the cascodes on the rev 2 board. the only way to be sure is to follow the traces and connections and compare them with the schematic. i'm looking at the schematic right now and i'm trying to think which transistors are which in the cascodes i'm thinking the cascode input transistors are a 3906 on the positive side and 3904 on the negative side, leaving the D01 as the common base part of the + side cascode (identified by the zener connected to the base) and the D51 on the - side (also with a zener connected to the base) i'm basing this from what i know for sure about the 3904 and 3906, that their max Vceo is far below the 150V total of the V++ and V-- rails. with the cascode stage, the 3906 and 3904 never see more than about 6V between E and C.
that is correct. (sorry about getting the 3904 and 3906 confused earlier, it's a blurry schematic)
Q9 and Q11 have the heat sinks. the cap that was above the board should have remained there, i think it was like that because of an artwork error in the solder side of the board.
most likely, the offset was caused by a bad cap or op amp in the servo circuit, and not the small signal transistors. there used to be an offset adjustment where a jumper was attached across the integrating cap and the offset pot adjusted. this was done at initial test or board test. i have a copy of the service manual, but it's in storage right now, so i don't remember the test points, i'm pretty sure that if you short across C15, the 100uf cap, you can adjust the offset pot. to test whether the servo is bad or the diff amp, you could short pin7 of the op amp to ground and see what the offset is like. if it's more than a volt or two, it might be the diff amp. if the offset gets worse when the short is removed from the output of the op amp, then it's the servo circuit. also make sure that the JFET discharges C15 when the power is shut off.
all of the boards should have the same rev number on them, if not then one or more of the boards have been swapped out in a repair. authorized servicers had access to the correct rev boards from APT if they needed them (while APT was in business), but i can see how boards might have got swapped out since then. btw, do you need a copy of the schematic? i can upload it if you need it
Q9 and Q11 have the heat sinks. the cap that was above the board should have remained there, i think it was like that because of an artwork error in the solder side of the board.
most likely, the offset was caused by a bad cap or op amp in the servo circuit, and not the small signal transistors. there used to be an offset adjustment where a jumper was attached across the integrating cap and the offset pot adjusted. this was done at initial test or board test. i have a copy of the service manual, but it's in storage right now, so i don't remember the test points, i'm pretty sure that if you short across C15, the 100uf cap, you can adjust the offset pot. to test whether the servo is bad or the diff amp, you could short pin7 of the op amp to ground and see what the offset is like. if it's more than a volt or two, it might be the diff amp. if the offset gets worse when the short is removed from the output of the op amp, then it's the servo circuit. also make sure that the JFET discharges C15 when the power is shut off.
all of the boards should have the same rev number on them, if not then one or more of the boards have been swapped out in a repair. authorized servicers had access to the correct rev boards from APT if they needed them (while APT was in business), but i can see how boards might have got swapped out since then. btw, do you need a copy of the schematic? i can upload it if you need it
Last edited:
Well, thanks again for your precious advice! I have a copy of the schematics from a pdf of the owner's manual that I downloaded from here, I would appreciate yours if it is higher resolution. Access to the repair manual would be great, as well as the history of the modifications/improvements done with the various revisions... if you could get them out of storage....
Tonite I plan to put the right transistors in the Q5 (06) and Q6 (51) positions in this par of the circuit (does it have a name?). Is this the best choice? Or should I use two 51 (sturdier?). Actually I am using A42 in place of 01 and A92 in place of 51, they should be equivalent (?). If it doesn't blow up again, I'll do the offset tests you reccommend.
Tonite I plan to put the right transistors in the Q5 (06) and Q6 (51) positions in this par of the circuit (does it have a name?). Is this the best choice? Or should I use two 51 (sturdier?). Actually I am using A42 in place of 01 and A92 in place of 51, they should be equivalent (?). If it doesn't blow up again, I'll do the offset tests you reccommend.
at the factory we had a dual banana plug with an 8ohm 2 watt carbon composition or wirewound resistor attached across the prongs. it was exactly the right distance apart to discharge from a fuse clip to the ground wire at the top edge of the amp board. you have to discharge both + and - rails. a 5 watt wirewound would be even better. carbon film and metal film resistors burn open when used for this, so don't use them.
having any transistors burn up prior to the drivers or predrivers when output devices shorted was a rare event
Q5 and 6 are a current mirror. you should keep the device types as you find them, since the 3906 and D51 were chosen for their particular places in the circuit for a reason. the 3906 only has a Vceo of 60V, but very high gain. the D51 (or the MPSA92) has a much higher Vceo that is well suited for where it is in the circuit, and probably has a lower gain. same goes for the complementary devices, don't mix up the NPN devices either between 3904 and D01 (or A42)
make sure you check the bias transistor and it's circuit very carefully. as with most bias transistor circuits, the transistor is best checked out of circuit. check the bias pot and make sure it's ok.and turned full counterclockwise before powering the amp up (use a dim bulb tester to protect against any further damage). setting the bias properly requires the use of a distortion analyzer and an oscilloscope. i have an alternate experimental method for setting the bias. email me for more info about it. it's experimental, so i don't want to explain it here and have somebody try it with the wrong equipment and let the magic smoke out
having any transistors burn up prior to the drivers or predrivers when output devices shorted was a rare event
Q5 and 6 are a current mirror. you should keep the device types as you find them, since the 3906 and D51 were chosen for their particular places in the circuit for a reason. the 3906 only has a Vceo of 60V, but very high gain. the D51 (or the MPSA92) has a much higher Vceo that is well suited for where it is in the circuit, and probably has a lower gain. same goes for the complementary devices, don't mix up the NPN devices either between 3904 and D01 (or A42)
make sure you check the bias transistor and it's circuit very carefully. as with most bias transistor circuits, the transistor is best checked out of circuit. check the bias pot and make sure it's ok.and turned full counterclockwise before powering the amp up (use a dim bulb tester to protect against any further damage). setting the bias properly requires the use of a distortion analyzer and an oscilloscope. i have an alternate experimental method for setting the bias. email me for more info about it. it's experimental, so i don't want to explain it here and have somebody try it with the wrong equipment and let the magic smoke out
Ok, I made some progress as I was able to put back the board in a way that it would not burn out! But I still have the original problem of a +~7 V on the output. It does not respond to the offset potentiometer setting. After lots of testing, I discovered that the problem originates from a + 7V that occurs on the part of the ground circuit that is connected to the signal input cable, the temperature resistor, the FET,, the 100uF cap etc. This offset is not on the ground circuit that is directly connected to the white ground wire. Looks like it is coming from the two cascaded diodes next to the offset potentiometer (by the way, this part of the circuit does not look like the schematics at all!) Upon further testing I discovered a resistance of ~2200 Ohm between that part of the ground circuit and the white wire. there is ) resistance on the good channel. I looked around a lot but I can't find how these two part of the circuit should be connected. I guess if I can eliminate this resistance the +7V will be sunk to ground and the offset hopefully will go away....
Thank you for any advice!
Thank you for any advice!