• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

DC OFFSET question

calman46

calman46

AK Subscriber
Subscriber
I have 2 Martex PM stereo 200 amps. (diy thread "increasing filter caps"). I have reacpped one amp and brought up on a DBT and all is good (no smoke). The DC offset on the right 60mV and the left is 300mV ,there are no trim pots. The input has three 2N4410 all common emitter and then two pairs of RCA 40595/40594 (these are antiques ). All these diode check good. The outputs are TIP35C/36C which are still installed and not tested but appear to be working . How should I check the idler current ? I'm looking at the resistors now as they are old maybe to far out of spec . If these amps are from the '70's they used old parts to build them.
Any suggestions ? thanks
 
Register to hide this ad
No ,I have found no infomation on these amps. There listed ( model numbers incorrectly listed I think ) in the '03 blue book with just manufactor dates '75-'78 . It is a dual mono design (two transfomers ). I left on the DBT to be safe .
 
Last edited:
They may test good, but if the amp uses a differential input, then the transistors much be matched to get close to 0-15mv of DC offset.

Can you test their HFE?

Rob
 
yes , the 2N4110 .Two of these must the differential pair and the lone one for the gain stage ? I'lll check the board again when I get home.
 
The HfE on 2n4410 are 73, 75 , 290. What about the wattage on the resistors in the picture ? There are 3 sizes I'm guessing 2w , 1w, and 0.5w ? I had 48Vdc on the new filter caps.
 

Attachments

  • IMG_0122[1].jpg
    IMG_0122[1].jpg
    97.4 KB · Views: 132
Across the bottom of your photo
Looks like yellow white black with silver 49ohms 10% 2W

Can't read the side of the square ones, but they should give the data on the side and look like 5W.

About center just above & between 2 square ones is RRO is 22K 10% 1W

Smaller ones are 1/2W mostly silver = 10%

Carbon resistors can change a LOT over time or not

The board is G10 glastic = good stuff

The hollow through board rivet solder points were before plated through hole. The rivet can loose contact to the trace if the solder job was not too good or has had enough thermal cycles. Use an ohmmeter to check and care if resoldering is necessary. Some solder paste (not acid) would help a bunch. If the contact is OK just leave them be.

The 1/2W resistor at the top (tipped can't see colors) of the board looks like it has a broken lead that would go to the 10 ohm 1/2W 10% at the top left.

Use a lot of care desoldering and don't if you can avoid it. Glastic held the traces the best in that time frame, but still come off easily

Observation only is that is quite the gain spread on the 2n4410
 
there 47 ohm (yellow ,violet ) so they are 2W . The white square are 0.22 ohm 5W.
yes the 22k is now 30k . I notice those through hole rivets,they seem ok but Ill recheck. I lifted those resistors to check them .
I checked the 2n4410 a few times I keep those hfe readings ,I ordered some new ones from Mouser.
I tried checking the idler current across one the 0.22ohm emitter resistors but I get a real small reading 0.05mv .
The transformers hum on both units but I have yet check if they just loose . I have not notice any over heating issues but I haven't left them on for long .
 
Last edited:
the gains on the 2n4410 you have are within spec on some of the old data sheets so you might not get any better ones.

What I get on the RCA 40594 & 5 are little T05 packages 2W 1A Ic. Seems small even for a driver.

Not sure where the .22 ohm 5W are in the circuit, but assuming they are emitter resistors for some higher power outputs your reading is not too far off the mark. Idle current of 50ma is not too uncommon for two parallel outputs. That would be 5mv per transistor emitter resistor.

Any idea when your amp was manufactured? I am not familiar with Martex amps so my comments are just general opinions.

Have you contacted Martex to see if you can get a schematic? Just a thought.
 
They were made in the mid 70's . There is not any kind of information available for these . I was thinking about a patent search. The amps do pass a signal its just DC offset is to high . I'll change out the differential pairs and put in new resistors and go from there .
 
I know it is a PIA, but the PCBs don't look that busy and right & left are almost identical. Looks like time to start making a schematic from the board.

It is not as bad as it seems once you get going and the 2 sides being almost the same helps getting it correct. Parts count is low which is a BIG help. I generally start in the center of paper so I can add where ever I need to. If it gets too messy I sometimes stop and redraw along the way.

The ones I have done start out really messy and get redrawn several times.

I am sure you have read at least part of EchoWars offset thread so know about matching gains of the Diff pair.

BTW nice looking / clean monster form the pictures you posted.
 
I put in two new 2n4410 for the differential pair ,hfe 160 . On power up the DC offset was -160mV(from 290mV) and It was still falling slow at -139mV when I turned off the amp(2-3 minutes on time). Its the transformers hum I'm concerned with both units have this hum.The transformers are mounted to the front of the chassis behind the face plate . I'm just not sure if something is straining them .Nothing gets hot and there all kinds of fuses on these amps it appears to normal. . I need to add an amp meter to my varrac . anyway
I'm all most done with the board schematic ,its my first time so .....
 
Transformer hum depends on how much varnish gets sucked into the lamination if it is VPIed or seeps in if just dipped. Lift an output lead and see how loud it is at no load. BTW a decent designed transformer runs at about 80 to 90% rated temp at no load. Most of the heating should be hysteresis and constant with the IR heating making up the rest.

AC ammeters are not as common as DC. I just use a Fluke on AC amps in series with the load if I care and an AC volt meter. There are LOTS of cheap DVMs with AC & DC amp scales that are accurate enough for most testing.

Mounting a transformer on a panel is a mechanical audio amplifier.

Output offset with no speakers connected draws almost no current. Hook up a hand full of carbon resistors that calculate between 4 and 8 ohms across the output terminals. They are cheap and if smoke then you need to find the problem. If they stay cool just keep looking for the cause of the offset.

BTW if you have a scope look at the output for oscillation. More common that you would think.

Be interested to see the schematic when you finish it. So far no one I know in Seattle has heard of Martek. Still a few left to hear from that were there in the 70s and into electronics.
 
FYI Its Martex. I changed a few more resistors and the third 2n4410.The dc offset now goes to -50mv in about 5 minutes but the other side is at -5mV. I was going to compare the voltage readings from one side to the other .This unit had a added chassis ground wire not original so that needs to be checked again ( also that rivet you mentioned).The schematic drawing is tough but I'll get it .
 
Here's a" diagram" of the input (differential amp) . I have another problem now ,the VU pins on turn on in the channel in question. I also noticed that one output transistors has a bad soldering iron burn .I will replace it maybe that is also effecting the offset. My TIP36c are not the same case type it maybe not fit heat sink. I'll deal that later.
 

Attachments

  • img013.jpg
    img013.jpg
    46.7 KB · Views: 38
I am guessing that you do not have speakers on the output so even though the VU pegged the amp dosen't draw much current.

I am also guessing you used higher gain Q1 & Q2 as replacements.

If you have a scope I bet you will see the output oscillating. At 100 to 200KHz of so. If no scope try your AC voltmeter It will probably give a high reading.

Haven't had much time to look at your drawing. Which is board 2 that the base of Q3 goes to.

Wouldn't worry much about the over zealous soldering of the TIP36C as long as it is otherwise OK.

50MV offset is not too bad. I have noticed most of my amps under test the offset goes down as the output into a dummy load increases. Maybe I am just blessed.

Are the 40595 & 40494 TO5 cases? The only data sheets I found showed them as that case and 2W rating. If they are the drivers for the TIPs the 2W rating on the data sheets I found is way low.

Sorry about the typo on Martex.

How about a picture of the 40595?

Do you have photo bucket? In the AK edit menu blue ball with the chain below it is for the link. In PB click the "direct" on the right side of the screen after you have clicked the picture and you will see a "Copied" briefly in yellow. In AK message edit menu bottom row clicking the blue ball will pop up "Please enter the URL of your link" and have "http:". If you just cntrl V it will paste the copied link there and the "http" will be replaced. Preview post will show you if you got it right. That puts the picture in your posting and doesn't use any of your AK data limit. If you already knew about it sorry for the rehash.

I was outside most of the day sunny and COLD! Not the best time for planing and whacking wood.
 
You must log in or register to reply here.
Back
Top Bottom