I have (15) BGW 750. They all work perfect except for 2. Hence, that is why I asked about the transistors. I have bridged these amps, and pushed them to THE ABSOLUTE MAX that I could safely without abuse. In all this time, only one took a **** on me and that is why I am doing some module work right now. The other one I picked up broke in order to both learn, and repair it at the same time. I may not be as experienced as the rest of you guys, but the BGW's are in my opinion some of the best sounding solid-state amps ever made. Remarkable that even today most 750 B & C models are 20-25 years old and are still cranked to the max daily. (750 A is a little older -introduced in '74). Djk, would you be willing to sell me some of the MJ15024/25 's? I am going to try this one module first and see how it comes out.
Since you're in GB, I would order the MJ21195/96 from Digi-Key. They have them in stock in one lot quantity for $3.78 each.
I'd leave the LM318 alone, it's not that bad. The input coupling cap, feedback cap, and the pair of level shift caps are what holds the sound back. Add a pair of 47µF 100V power supply bypass caps across the main filter caps.
Don't mix output transistor types, replace all in a channel. If you want good reliability, do both channels.
I'd leave the LM318 alone, it's not that bad. The input coupling cap, feedback cap, and the pair of level shift caps are what holds the sound back. Add a pair of 47µF 100V power supply bypass caps across the main filter caps.
Don't mix output transistor types, replace all in a channel. If you want good reliability, do both channels.
You can draw your own schematic out.
Opamp runs on ±15V derived from zeners off the main rails.
The output of the opamp goes through a series pair of resistors to the positive rail, and a series pair of resistors to the negative rail.
The Vas pair are driven by these resistor pairs, and the resistors connected to the opamp output are bypassed with a largish electrolytic.
Vbe multiplier between the Vas pair.
Tripple EF output stage.
Power supply, relay driver, and crowbar.
Fini.
Opamp runs on ±15V derived from zeners off the main rails.
The output of the opamp goes through a series pair of resistors to the positive rail, and a series pair of resistors to the negative rail.
The Vas pair are driven by these resistor pairs, and the resistors connected to the opamp output are bypassed with a largish electrolytic.
Vbe multiplier between the Vas pair.
Tripple EF output stage.
Power supply, relay driver, and crowbar.
Fini.
NTE has a sub for it the NTE918M. Its is probably just a re-labeled LM318 op amp. Just look it up to be sure its the metal can variety which is what you will need for the BGW. NTE has alot of antique replacement parts. I bought an ua709 NTE sub today for a Wavetek Function generator that was made in 1969! Works perfectly.
Mark
Mark
So far the only LM318 I can obtain is the 8 pin dil version and not the To39 can
Is there a problem using this ?
Does any one have a metal can unit they are willing to sell /supply
I will try and contact NTE but trust me its a bank holiday here in the UK
I must say the construction of the BGW is a bit of a suprise with all the components being mounted on the track side its a bit like the old blob boards I used years ago! Its as if the components were thrown at the board does any one have thoughts re the construction
regards Trev
Is there a problem using this ?
Does any one have a metal can unit they are willing to sell /supply
I will try and contact NTE but trust me its a bank holiday here in the UK
I must say the construction of the BGW is a bit of a suprise with all the components being mounted on the track side its a bit like the old blob boards I used years ago! Its as if the components were thrown at the board does any one have thoughts re the construction
regards Trev
Hi Trev,
Don't change the op amp yet. There is a procedure for balancing the driver currents as you monitor the DC offset of the 318. I'll bet you have not done that. If I only still had my manuals!!!!
On the board near the MJE devices you will see resistors with pads (or resistors) beside them. One adjusts bias current, the other two set the currents in the drivers. The spare resistor pads the value of the main one. Start by removing the pad and see if the offset gets worse, or better. Now you know which way to go. Make sure you aren't playing with the bias current! Adjust until the offset on pin 6 is below 100mV (as I recall).
I have never seen the op amp bad on these. I have seen an op amp supply capacitor short and create offset that way.
-Chris
Edit: Mark, I try to avoid NTE / ECG garbage unless there is no other way. That stuff is expensive trash.
Don't change the op amp yet. There is a procedure for balancing the driver currents as you monitor the DC offset of the 318. I'll bet you have not done that. If I only still had my manuals!!!!
On the board near the MJE devices you will see resistors with pads (or resistors) beside them. One adjusts bias current, the other two set the currents in the drivers. The spare resistor pads the value of the main one. Start by removing the pad and see if the offset gets worse, or better. Now you know which way to go. Make sure you aren't playing with the bias current! Adjust until the offset on pin 6 is below 100mV (as I recall).
I have never seen the op amp bad on these. I have seen an op amp supply capacitor short and create offset that way.
-Chris
Edit: Mark, I try to avoid NTE / ECG garbage unless there is no other way. That stuff is expensive trash.
I have replaced the lm318 with a plastic cased version
offset now down to 0.8 mv all seems fine.
Output transistors have been replaced with mj 15024/25 devices and that seems to be ok also.
The opamp rails were measured at 10 volts each not 15 as earlier stated !!!
The amp is rather primative with its ab-b output stage and I wonder what the transfer function would look like!
Seems to me it one of those high feedback designs that relies on its feed back to obtain any thing like a reasonable performance rather than being a highly linear design.
Allthough i do like the method of obtaining a symetrical VAS stage
Thanks for your help etc
regards Trev
offset now down to 0.8 mv all seems fine.
Output transistors have been replaced with mj 15024/25 devices and that seems to be ok also.
The opamp rails were measured at 10 volts each not 15 as earlier stated !!!
The amp is rather primative with its ab-b output stage and I wonder what the transfer function would look like!
Seems to me it one of those high feedback designs that relies on its feed back to obtain any thing like a reasonable performance rather than being a highly linear design.
Allthough i do like the method of obtaining a symetrical VAS stage
Thanks for your help etc
regards Trev
Symmetry in BW750.
If something looks symmetrical it sure looks symmetrical, but on the other hand is it symmetrial? Are the electrons moving about in the cirquits experiating a symmetrical environment. Since electron mobility is differing a factor of 3 in N resp P doped Si, the symmetry is apparent and only on the schematic. To get symmetry in true sence is to use class A.
Andreas
BTW BGW750 is the worst sounding power amp I've heard to date.
If something looks symmetrical it sure looks symmetrical, but on the other hand is it symmetrial? Are the electrons moving about in the cirquits experiating a symmetrical environment. Since electron mobility is differing a factor of 3 in N resp P doped Si, the symmetry is apparent and only on the schematic. To get symmetry in true sence is to use class A.
Andreas
BTW BGW750 is the worst sounding power amp I've heard to date.