Geller Labs
Geller Labs


Aeroflex Weinschel 8310 for the Electronics Hobbyist

A tech note in progress ... revised 3/10/2014

(use your browser's refresh function to be sure to see the latest changes! )

There are many experiments where an amateur scientist, electronics hobbyist, or student needs a very small signal source, often of lower amplitude than can be had with a bench signal source. For example, for some of my recent professional work and now as an amateur scientist / electronics hobbyist, I needed micro volt audio sources as input to amplifiers being designed for noise measurements. Also, I needed a calibrated 2 kHz signal to test, align, and verify the proton magnetometer experiment. Another example was my work in understanding the low end scale response of a DMM's AC voltage range.

For all of these experiments, I used a Kay 837 or a Kay 839 manual attenuator (or, on ocassion, both in series). The Kay attenuators are very high quality, easy to use manual attenuators (highly recommended additions for the amateur electronics laboratory).

However, there are two potential improvements.

First, it is very handy to have an attenuator that can be adjusted up or down in steps without having to mentally calculate combinations of switches to set a particular value.

Second, for those fortunate to have some means of computer control, a computer controlled step attenuator can be useful in all sorts of automated test set ups.

I first came across the Aeroflex Weinschel 8310 in an eBay sale of a new old stock unit with the manual. It was configured as a 75 ohm attenuator, of limited use for all but perhaps those working in radio, cable TV, and TV. In my amateur science lab, just about every experiment is based on 50 ohm terminations. Yet, at $150, almost "new", that was too hard to pass up. The unit arrived, in every bit of pristine condition as promised (said to have been left over from an abandoned project of years past).

I was pleasantly surprised on inspection to realize that the unit is based on the Aeroflex Weinschel 8210A controller. While the 8310 manual offers little guidance as to configuration, the 8210A manual tells almost all you need to know.

The 8210A attenuator systems are set up to communicate with one or more attenuators over a Weinschel proprietary bus. Only certain model variations accept the IDC connector and communicate with the 8210A protocol (more later).

It turns out that many of these compatible attenuators are also sold surplus (often as "new old stock" surplus) from time to time at forums like eBay. Often such finds are sold in groups of 2 to 10 pieces and seem to become available from military surplus sales, bankruptcies, and completed or failed small business projects.

Bottom line is that beautiful lab quality gear from the 8210A and 8310 related attenuator families that cost many thousands new can be often be had at very reasonable prices as surplus gear. It is not all that surprising, as amateurs and hobbyists have been buying many thousands of units of relatively high end gear at forums ranging from hamfests to eBay for many years. Some higher frequency versions typically sell at much higher prices. Also, commercial used prices are far higher.

When I first started learning about / experimenting with the 8310 and related compatible bus attenuators (usually, but not always the "-T" models), there was a variety to be had at very reasonable surplus prices. For now, it might actually be that I bought all the stragglers that had been out there waiting from some hobbyist interest. However, as all surplus markets go, what is dry today could be any number and variety of surplus 8310 and/or attenutors in the near future. The less desirable 75 ohm 8310 either one or two channels is probably the best entry method to pick up the chassis with the 8210A, comms interface, power supply, etc, then hunt for one or more compatible surplus attentuators for your needs. On the other hand if you ultimately want a 50 ohm unit, starting with a 50 ohm unit might yeld a high quality set of internal shielded cables and 50 ohm panel ports.

If you are a commercial / military operation who happened on this web page by a search, easy answer, contact Aeroflex, Wienschel, and specify and buy a brand new unit configured to your needs. Then, in five to twenty years, hopefully your unit might end up on the surplus market for its second life with an amateur!

The 8310 is nicely packaged in an aluminum box with a front panel with LED numeric display and buttons. With a 8310 surplus buy, you get a power supply, an 8210A controller (both serial and GPIB, possibly USB if you get a very late model version), one to four bus controlled attenuators, connecting cables, and some sort of RF panel connection, typically BNC, SMA, or an odd flat microwave threaded assembly that often terminates in a standard N connector. It does not appear to be standard a AP7 microwave connector, but is very similar in layout with locator tabs around the center.

The good news is that by use of the 8210A manual, you can program a unit for any compatible bus attenuators. They come in all sorts of varieties with various numbers of steps, size of the minimum steps, frequency ranges, etc. A really interesting 8210A feature is that you can program "groups" having two or more attenuators where you just send the desired total attenuation, and the 8210A controls the group as if it is one single attenuator (of course to use this method, you connect the attenuator inputs and outputs in cascade (series) either internally or by the front or rear panel connectors).

I wanted to make my first single channel 75 ohm unit into a two channel 50 ohm unit. One attenuator that I found is about 64 dB in 1/4 dB steps. Another is 32 dB in one dB steps. I wanted a two channel unit with the option of using the attenuators as separate attenuators (each with its own BNC input and output) or cascaded as a group of two series attenuators. Using the 8210A manual, setup (with some learning curve for the odd syntax) went fine. (Programming examples to follow). The problem is my like new case and display are limited to one channel under front panel control (GPIB control doesn't care about the front panel programming and works fine by the 8210A manual configuration).

Programming example:

ASSIGN AT1 3200T-2 57830
ASSIGN AT2 3201T-1 1896



check after programming (*RST or cycle power):

2, AT1, 3200T-2, 57830, 2, AT2, 3201T-1, 1896, 4

Note that the type attenuator, e.g. 3200T-2 does not literally include the quotes from the 8210A manual example. '3200T-2' (There might be some acceptable quote " or '; I have not yet got it to work that way.)

Until you get the programming correct for your reconfigured unit, the display will read *FAIL* after complete power up.

More to follow ... presently off on other projects, will update later with revised programming including programming the front panel display.

Plan is to build a 8310 unit with a 127 dB 1dB step attenuator (3200T-1) and a 64 dB 1/4 dB step attenuator for a combined wide range, 1/4 db resolution CH 3 virtual attenuator (the cascade of CH1 and CH2).

The new combination works extremely well. Channel 1 (CHAN1) is a 127 dB, 1 dB step attenuator between Ports A and B (microwave connection to N to BNC). Channel 2 (CHAN2) is a 63.75 dB, 0.25 dB step attenuator between Ports C and D. And, Channel 3 (CHAN3) is a virtual combined attenuator, the sum of the two 193.75 dB with 1/4 dB steps between Ports A and D (with B and C coupled in cascade). (very cool!)

Pictures and programming instructions to follow.

Here are the programming lines that I used over GPIB:

query LIST? DEVICE CONFIG returns: 2, NONAME1, 3200T-1, 1, 2, NONAME2, 3200T-2, 57830, 4
(check the hardware on the attenuator bus, the numbers following the model numbers are the serial numbers programmed into each of the attenuators)

ASSIGN AT1 3200T-1 1
ASSIGN AT2 3200T-2 57830

query LIST? DEVICE CONFIG returns: 2, AT1, 3200T-1, 1, 2, AT2, 3200T-2, 57830, 4
(Check the programming to confirm that the attenuators are named as desired. If you get a NONAME, stop and keep working until all of your attenuators have the desired name.)

ASSIGN CHAN 1 ATTN AT1 LED 0x31 (lights the first two port LEDs A and B, by 3=0011; and the CH1 LED)
ASSIGN CHAN 2 ATTN AT2 LED 0xC2 (lights the last two port LEDs C and D, by C=1100; and the CH2 LED)

ASSIGN CHAN 3 ATTN CHAN3 LED 0x94 (lights the first and the last port LEDs A and D, by 9=1001; and the CH3 LED)

The LED HEX code (8 bit byte) first nibble following the x is the binary code equivalent (4 bit nibble) for the Port Lights, the last 4 bit nibble determines which of the four channel lights to illuminate. Very clever, and easy to program once you understand the sytax and programing method. The CH numbers are determined by the binary nibble 1 bit position: CH 1, 1 (0001); CH2, 2 (0010); CH3, 4 (0100); CH4, 8 (1000). The bit position of the "1" determines which channel light LED will illuminate.

If you want to be begin your new configuration from a fresh start, you can send ERASE ASSIGN. However, don't do that until you are sure you understand the programming method or you might brick your unit until you come up to speed on 8310 / 8210A programming.

I have no affiliation with Aeroflex Weinschel, nor take any responsibility for your 8310 / 8210A experiments. If you are lucky to find one of these units on the surplus market at a reasonable price, I highly recommend it as a great addition to a high end amateur science or electronics hobbyist shop / laboratory.






Tech Notes
About Geller Labs