Resurrecting a 1980s analogue preamp project

[thatmetalman]

1 hour ago, Leinster Lad said:

Had a little look at the 5000-pre that was not working and discovered that the BC640 in the power supply positive rail was open.

Had a set of BD681 /682  (darlingtons) so with a bit of effort to get access to the copper side of the main board, I got the old 639 / 640 out and the 681/682 fitted.

Drop straight in as the pin out was the same.

 

Fired it up and bingo !     back to life.

Bit of a soak test and they are still at room temp

Measured 29.4V across the 4.7K base resistor, so about 6mA thru the bases.

VEC on both is about 0.64V  so they are well and truly turned on.

 

The 681/682 are rated for  4amp nominal collector current, peaking at 6amps  so they should be  all good.

Hi David.

 

That's a great report. I'm going to swap out the BC639/640 for the same type you've used. Readily available. Bill sent through a bunch of good suggestions as you saw, so I'm working on those right now. Will make the Darlington changes while I'm at it (I was never 100% comfy with the BC640 even though I knew it was 1A rated and not passing anywhere near that current).

 

Cheers,

 

Tim

[Leinster Lad]

No worries Tim, happy to share. I decide on using the darlingtons because of their low power requirements to turn on and their relative high current handling.

 

So while sitting here pondering, I got to thinking about the 400hz onboard oscillator.

I was sure I read back in the day about setting the output via RV4 to 0dB, so that the master level could be set so the VU meters would also read 0.

 

Cannot for the life of me find that reference  so i'm thinking of setting the output of the oscillator to 1.0vrms which I think should be 0dB ?

 

If I then calibrate the meters to read 0 ( first Red LED ) at 1.0vrms  ( from the 400hz osc ) then the planets should all align and the vu level should at least be fairly accurate.

 

Thoughts ?

[Leinster Lad]

Hmmm,  measuring at the trim pot I can only get about 

 

At the opamp output ( after C24   )  I can get  376Hz @ 372mV RMS ( 1.09V p2p )

 

I'm not sure if 0dB is  1V p2p or RMS ?

[betocool]

Depends on the dB! dB is nothing more than the relationship. So, 0 dB is 100% or a factor 1. Usually the dB is followed by a letter, so, dBm refers to milliwats (0 dB being 1 mWatt). From my understanding, 0 dBV is 1 V RMS, and that is 2.8V pp. As a rule of thumb, -3 dB is the equivalent of 70% (roughly), -9 dB is 0.35 or 35%, and that is about 1Vpp from 2.8Vpp.

 

What you are showing in the scope, 1V pp is about -9 dBV.

 

Vpp to VRms is a factor 2.8.

 

The graph below gives a rough idea of dBV vs V vs dBu:

 

Hope it helps.

 

Cheers,

 

Alberto

[Leinster Lad]

Outstanding Alberto !  Exactly what I need !

The setup procedure for the series 5000 pre-amp and meters is strangely lacking in this important detail.

 

Looking at the block diagram for the series 5000 pre-amp and I noticed that the tape 1 & 2 inputs connect directly to the VU meters via the tape monitor switch !

 

So I connect my trust CD player with a test disc inserted, to the "tape 1" input.  switch the monitor switch over to "tape 1"  and play a 400Hz test tone at 0dB  !!

 

Track 10 is specifically for testing VU meters.  Gotta say, these LED meters are really really accurate when displaying the various pulse tests. 

100ms, 10ms, and 1ms pulses are displayed at exactly 0dB

I am very Impressed !

 

Meters are pretty close, but i will let everything warm up for a bit and then do the final "sensitivity" adjustment (RV1) on the meter boards to display 0dB  !!

(currently the left (top) meter is about 0.5dB low with a constant 400Hz 0dB tone.

 

Sorted !!

 

Using this test disc  https://www.discogs.com/release/1607502-Unknown-Artist-Test-Compact-Disc-Extensive-Tests-For-Audio-Equipment

[thatmetalman]

On 09/11/2018 at 9:00 PM, Monkeyboi said:

RV5 and RV6 are for the d.c. offset adjustments for the op amps in the monitor output stage.  See the attached schematic. 

 

To successfully adjust the d.c. offset to allow a capacitiveless coupling of the output you need to have access to a d.c. microvolt meter to affect the adjustment or a high sensitivity CRO.  It imperative that the d.c.voltage at pin 6 of each op amp be exactly 0 volts under no signal conditions after adjustment of RV5 and RV6 for each channel independently.  Any d.c present at the output will cause damage to the speakers connected to a fully d.c. coupled power amplifier, hence the adjustments are critical.  If any of the components in this part of the circuit are replaced or the power supply repaired of modified, the d.c. offset adjustment(s) must be checked prior to placing the pre-amp back into service.

 

The links LK1 and LK2 allow two possible modes of operation.  When the shorting links are in position A this keeps the output capacitor in each channel in circuit for a.c,.coupling.  In position D it facilitates d.c. coupling of the output by shorting out (bypassing) the output capacitor in each channel.

 

Values for RV5 and RV6 are 100k 25 turn cermet trimpots.  I used the Radytronic ^TM 2396W style vertically mounted units because of space constraints, ease of adjustment and its local availability.

R92 and R93 are 22k 0.25W metal film resistors.  I mounted them on the underside of the board.

 

If you have already commenced the build of the main board you will have to replace resistors R82 and R83 (6k8 025W metal film) as these need to the slightly physically located on the board to facilitate the modification.

 

Cheers,

Alan R.

 

DISCLAIMER:  This modification is still under test.  Your decision to implement this modification is at your own risk and I accept no liability for the suitability or long term performance of the pre-amplifier as a result.  I also have no control over the quality of the workmanship nor the testing ability of the person(s) undertaking this modification.  

 

Hi @Monkeyboi Alan

 

I have a question about your AC/DC coupling mod. My understanding is that most accessible power amps are AC coupled. Is there a scenario where DC coupled is preferred and can be used on any power amp (presuming correct set up to 0V DC)? Or is a DC coupled output only suitable for power amps expecting capacitiveless in? Is the DC adjustment via each trimpot only necessary when DC coupled is selected?

 

Cheers,

 

Tim

 

Edited August 30 by thatmetalman

[thatmetalman]

5 hours ago, Leinster Lad said:

Outstanding Alberto !  Exactly what I need !

The setup procedure for the series 5000 pre-amp and meters is strangely lacking in this important detail.

 

Looking at the block diagram for the series 5000 pre-amp and I noticed that the tape 1 & 2 inputs connect directly to the VU meters via the tape monitor switch !

 

So I connect my trust CD player with a test disc inserted, to the "tape 1" input.  switch the monitor switch over to "tape 1"  and play a 400Hz test tone at 0dB  !!

 

Track 10 is specifically for testing VU meters.  Gotta say, these LED meters are really really accurate when displaying the various pulse tests. 

100ms, 10ms, and 1ms pulses are displayed at exactly 0dB

I am very Impressed !

 

Meters are pretty close, but i will let everything warm up for a bit and then do the final "sensitivity" adjustment (RV1) on the meter boards to display 0dB  !!

(currently the left (top) meter is about 0.5dB low with a constant 400Hz 0dB tone.

 

Sorted !!

 

Using this test disc  https://www.discogs.com/release/1607502-Unknown-Artist-Test-Compact-Disc-Extensive-Tests-For-Audio-Equipment

G'day.

 

Well, the setup was omitted from the 1981 articles. Just found out yesterday. But it seems you got sorted!

 

The setup actually appeared within an errata published in ETI December 1981. Here it is copy paste from page 12 of that month:

 

"The 400 Hz oscillator procedure was omitted. It’s simple. Take your multimeter, set to read AC volts, and connect it between the wiper of RV4 and 0 V. Set RV4 to obtain 1.2 VAC (RMS).
Note that R52 and R53 on the overlay are shown as 220R when they should be 220k as per the circuit and parts list."

 

The LED calibration/setup instructions were published in OCT 1981 (part 3 of 3 of the 5000 preamp article), and of course the LED original article from June 1981. Do you have those? I have the setup printed on the LED meters PCB.

[thatmetalman]

5 hours ago, Leinster Lad said:

Outstanding Alberto !  Exactly what I need !

The setup procedure for the series 5000 pre-amp and meters is strangely lacking in this important detail.

 

I can't remember who posted the Archive link to all 1981 PDFs, but here it is again anyway: https://archive.org/details/ETIA1981/

 

[Leinster Lad]

18 minutes ago, thatmetalman said:

G'day.

 

Well, the setup was omitted from the 1981 articles. Just found out yesterday. But it seems you got sorted!

 

The setup actually appeared within an errata published in ETI December 1981. Here it is copy paste from page 12 of that month:

 

"The 400 Hz oscillator procedure was omitted. It’s simple. Take your multimeter, set to read AC volts, and connect it between the wiper of RV4 and 0 V. Set RV4 to obtain 1.2 VAC (RMS).
Note that R52 and R53 on the overlay are shown as 220R when they should be 220k as per the circuit and parts list."

 

The LED calibration/setup instructions were published in OCT 1981 (part 3 of 3 of the 5000 preamp article), and of course the LED original article from June 1981. Do you have those? I have the setup printed on the LED meters PCB.

Thanks Tim,

Yes I have the Dec 1981 ETI.

Pretty sure I'm not able to get more than about 0.2 AC RMS  out of the oscillator. not sure exactly why yet....

Doesn't really matter anyway, as you can see, i've developed a cunning plan !

[Wicked_XR]

On 27/08/2024 at 6:19 PM, Leinster Lad said:

Had a little look at the 5000-pre that was not working and discovered that the BC640 in the power supply positive rail was open.

Had a set of BD681 /682  (darlingtons) so with a bit of effort to get access to the copper side of the main board, I got the old 639 / 640 out and the 681/682 fitted.

Drop straight in as the pin out was the same.

 

Fired it up and bingo !     back to life.

Bit of a soak test and they are still at room temp

Measured 29.4V across the 4.7K base resistor, so about 6mA thru the bases.

VEC on both is about 0.64V  so they are well and truly turned on.

 

The 681/682 are rated for  4amp nominal collector current, peaking at 6amps  so they should be  all good.

I would get rid of the tags, they like to smoke up. I turned my preamp on after 10 years, and the tag next to the VU meter reg lit up...I would also put a heatsink on the VU reg..

 

[Leinster Lad]

1 hour ago, Wicked_XR said:

I would get rid of the tags, they like to smoke up. I turned my preamp on after 10 years, and the tag next to the VU meter reg lit up...I would also put a heatsink on the VU reg..

 

But replace them with what ?

They were specified as TAG's for a reason, i just don't know what the reason was ?  Can normal electro's do the job as well ?

[Leinster Lad]

Okay, just a little more tinkering.

I did notice that the R5 4.7k resistor at the bases of the BD681/682 darlingtons that I fitted does get a little warm.

Old ohms law tells me that it has about 6mA going thru it, which equates to about 200mW. 

My board as well as R5 show signs of elevated temperatures.

 

So I dabbled with a simulator and had a play with different R5 resistors to see what effect that had on the trannies and the base resistor.

Those darlingtons sure are sensitive buggers !

 

In fact, even with the resistor at 33K  the Vec only increase from 0.896v ( 4.7K 6.2mA 28.93v 0.178W ) or 0.918v ( 10K 2.89mA 28.932v  0.084W ) 

to 0.988v ( 33K  0.878mA  28.934v  0.025W )

 

Looks like a 10K resistor will work real nice !  trannies near fully conducting and bugger all base current !

Edited August 30 by Leinster Lad

[Wicked_XR]

On 27/08/2024 at 6:19 PM, Leinster Lad said:

Had a little look at the 5000-pre that was not working and discovered that the BC640 in the power supply positive rail was open.

Had a set of BD681 /682  (darlingtons) so with a bit of effort to get access to the copper side of the main board, I got the old 639 / 640 out and the 681/682 fitted.

Drop straight in as the pin out was the same.

 

Fired it up and bingo !     back to life.

Bit of a soak test and they are still at room temp

Measured 29.4V across the 4.7K base resistor, so about 6mA thru the bases.

VEC on both is about 0.64V  so they are well and truly turned on.

 

The 681/682 are rated for  4amp nominal collector current, peaking at 6amps  so they should be  all good.

I would get rid of the tags, they like to smoke up. I turned my preamp on after 10 years, and the tag next to the VU meter reg lit up...I would also put a heatsink on the VU reg..

 

[Wicked_XR]

5 hours ago, Leinster Lad said:

But replace them with what ?

They were specified as TAG's for a reason, i just don't know what the reason was ?  Can normal electro's do the job as well ?

Replace it with a 100uf, and put a 0.22uf mono direct on the reg output, the reg does need a heatsink, even if it doesn't sit flat on PCB, it is getting hot. I've dropped a couple generic transistors to replace the BDs, and used the PCB with a bit of heatsink. They do get warm. 

[thatmetalman]

Gents,

 

Some progress following your suggestions.

• 25v AC inputs amended to 15v per rail
• LED 7815 regulator now with 0.1uF mono cap across output and 0v
• LED 7815 regulator now with 100uF 25v electro in place of tag tant.
• LED 7815 reg GND tab added to PCB 
• Q1 and Q2 BC639/BC640 pair now BD681/BD682; ergo,
• R5 changed to 10k
• Copper re-routed away from regulator heatsinks
• Ground pour added to beneath all regulators
• Var. regs reorientated for symmetry
• PSU setup test points moved to align with their respective regs
• Text added for original magazine article references. This was done to assist any future servicing by 3rd parties.
• Monitor Amp out reference text amended to reflect DC / AC settings (TBC regarding 'DC only' comment accuracy)
• 400Hz OSC setting text added in line with DEC 1981 errara notes

I've been fluffing about with the original ETI logo in Photoshop. I might add that as some candy towards the end. Personally I think Tilbrook's face from the ETI front cover would be better 

 

Tim

 

 

 

 

 

 

 

Edited August 30 by thatmetalman

[Leinster Lad]

Outstanding work Tim

Would be honoured to revamp my second series 5000 pre-amp with updated boards and parts.

My general knowledge base has increase heaps since I built it originally, and I'm sure I'd do a better job the second time around !

( and I have better test equipment as well )

 

Please let me know if/when a set of boards will be available to purchase from you 

 

Cheers

Dave

[thatmetalman]

Hi Dave.

 

Thanks for the kind words. I'm literately working on the sub boards now, increasing the font sizes (which forces moving text around). Quite a rework actually. Hope to have it done soon.

[Leinster Lad]

Excellent, and yes, I am well impressed with your efforts !

[Leinster Lad]

Hi Tim,   noticed that R31 and D12 are transposed.  All other mute trannies have the diode connected to the base.

I don't know if this actually  electrically make a difference, and I know this is how the OG circuit is, but since you are going to so much effort.......

 

 

Edited August 30 by Leinster Lad
spelling

[thatmetalman]

Hi Dave.

 

Good pick up! It is how the 1981 PCB is configured, i.e. my CCT reproduction mirrors the board.

 

To confuse matters further, the CCT on page 34 of ETI OCT 1981 shows R32 and D13 swapped (but D12 and R31 shown as expected). The PCB reflects the opposite for each pair.

 

I checked the 1981 PCB layout yet again just now, and it does reflect the PCB layout on page 38 of the original mag.

 

I would say that the errata was never picked up unless it appeared early 1982, which I don't have access to check.

 

There are/were many, many hundreds of the preamp successfully constructed (as you know!). I'll mirror the board unless someone such as Bill or perhaps Alberto want to step in and give the all-clear to swap R31 and D12. I suspect that such a change would be fine.

 

Cheers.

[Leinster Lad]

I figured that swapping would be fine as it would then align with all the other mute transistor circuits.

I can see no reason why just one would be different, and my electronic theory knowledge is not strong enough to determine IF it would actually matter !

 

Hopefully one of the gurus will chime in soon

 

@betocool Alberto  @Monkeyboi  Alan  

[thatmetalman]

In the meantime, here's the change. Sometimes you gotta' go with your gut...

 

[Leinster Lad]

100% it will be fine.

 

I am still interested to know the actual "electrical" effect swapping the diode and resistor will have ......   ( pretty sure it is zero, but would be good for an actual qualified expert to verify )

[betocool]

In that configuration, having the diode and the resistor either way will work the same. That is usually the case with components in series. Most important is the direction of the diode!

 

Cheers,

 

Alberto

[Wicked_XR]

asked a few people about the diode/res swap, not an issue.....no one can think of a reason why.......its sort of same as resistor on a led, if you put it on cathode or diode, does the same thing